Re: what is an "in-core" disklabel ?
On 2012.10.08 18:22, Robert Bonomi wrote: > 'cached' is not _technically_ exactly accurate, but you have the concept > basically correct. Thanks for the detailed explanation, Robert. Maybe "shadowed" would be have been a more accurate term. But "in-core" also has a nice ring to it! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: what is an "in-core" disklabel ?
> Date: Mon, 08 Oct 2012 17:14:20 +0200 > From: "Lucas B. Cohen" > Subject: what is an "in-core" disklabel ? > > Hi, > > I've seen the term "in-core" a couple times while reading up about BSD > disk labels. Does it refer to data that is cached in kernel memory ? > > Context examples : > > - fdisk(8) outputs "parameters extracted from in-core disklabel" > > - bsdlabel(8)'s manual explains that the -n (dry run) parameter "does > not install the new label either in-core or on-disk". 'cached' is not _technically_ exactly accurate, but you have the concept basically correct. The O/S reads the label information and stores it in an internal data structure, Then, when it needs to use that data (frequently!:) it uses the values in that internal structure, rather than attempting to re-read from the disk, itself. Technically, it's _not_ "cached" -- cached data is used to short-circuit a 'read' attempt, using an in-memory block of byte instead of an actual disk transfer. The -effect- is similar, but there are *important* differences. 'Cache' data is integrated with I/O operations, and a _write_ to the place where the data was read from -invalidates- the cached data, whereupon, the next read attempt will *not* be short-circuited, and the actual on-disk data will be returned. In the case of the disk label, it is read (once) into the internal data structure, and only the internal data is used after that. A userland app can change the 'on disk' data -- or trash it completely -- and what the O/S "thinks" the label info is will NOT be affected by that change to the 'on disk' data. The warnings you see in the documentation, are reminders that the O/S's 'internal' data and the 'on disk' data are *NOT* necessarily the same. That looking at _one_ source of that data does *not* guarantee that what you see =there= is the same as what is in the other place. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
what is an "in-core" disklabel ?
Hi, I've seen the term "in-core" a couple times while reading up about BSD disk labels. Does it refer to data that is cached in kernel memory ? Context examples : - fdisk(8) outputs "parameters extracted from in-core disklabel" - bsdlabel(8)'s manual explains that the -n (dry run) parameter "does not install the new label either in-core or on-disk". ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: fdisk/bsdlabel/disklabel: Class not found?
On Thu, Dec 10, 2009 at 09:23:30PM -0500, Nick Dalsheimer wrote: > So we are saying, that bsdlabel and fdisk are broken? This is *very* > disappointing. Huh? > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: fdisk/bsdlabel/disklabel: Class not found?
So we are saying, that bsdlabel and fdisk are broken? This is *very* disappointing. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: fdisk/bsdlabel/disklabel: Class not found?
bsdlabel: Class not found re-edit the label? [y]: You cant edit it. You can only say "N" and it exits w/o saving any changes. This is very annoying, because you cannot do anything with the label unlike the old days.. I had to mount an older drive and then I was able to edit the bsdlabel on the 8.0 drive as it was not 'online' - -- J.D. Bronson Information Technology Aurora Health Care - Milwaukee WI ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
fdisk/bsdlabel/disklabel: Class not found?
Hello all! I've found many references to this error on Google: "fdisk: Class not found, bsdlabel: Class not found" but none explain what this error *means*. Could someone explain this error and possible remedies? I'm using a custom 8.0-RELEASE-p1 kernel. I don't even need to edit the label in order for bsdlabel to spit this out! Any information would be very helpful! Thanks, Nick ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: disklabel output format ? How to see in G M ..
On Thu, Apr 30, 2009 at 08:51:13PM +0300, Anonymous wrote: > > Using disklabel -A /dev/da0s1 I would like to see the sizes in G or M > format, how can I do this? > Also, googling arround i found output showing the cylinder space occupied > by a partition (like : > # cyl* X - Y ). How do I see that ? > PS: i did man disklabel and bsdlabel but i didnt find the correct > arguments. > thank you. I don't know if it will display them that way, but you can enter them as 100M or 12G or whatever is appropriate when you are creating partitions. jerry > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: disklabel output format ? How to see in G M ..
On Thu, Apr 30, 2009 at 12:51 PM, Anonymous wrote: > Also, googling arround i found output showing the cylinder space occupied > by a partition (like : > # cyl* X - Y ). How do I see that ? I think that fdisk will show you this. Good luck-- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
disklabel output format ? How to see in G M ..
Using disklabel -A /dev/da0s1 I would like to see the sizes in G or M format, how can I do this? Also, googling arround i found output showing the cylinder space occupied by a partition (like : # cyl* X - Y ). How do I see that ? PS: i did man disklabel and bsdlabel but i didnt find the correct arguments. thank you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: "disklabel: Class not found" when editing USB disk
Canonical way to boot FreeBSD by hand, with no tools. Using one slice, one partition fdisk -BI /dev/da0 bsdlabel -Bw /dev/da0s1 newfs /dev/da0s1a Making sure there's a /boot directory with the all important loader Making sure /boot/kernel/kernel exists This should boot -- however is pretty pointless without anything else. init's needed, and everything else to make a full MUM (Multi User Mode) environment... Keep note that not only does the MBR need bootable code, so does the bsdlabel/disklabel newfs will set the proper fstype in the bsdlabel if it's partition is 'a' Good luck... and yes, I top posted. (Deal with it, people who don't like it...) On Thu, Feb 26, 2009 at 10:30 AM, Michael W. Lucas < mwlu...@blackhelicopters.org> wrote: > Hi, > > I'm following Ceri's instructions for getting FreeBSD on a flash > drive. My laptop is a Feb 1 -current, I'm installing FreeBSD 7.1, and > the instructions are for 6.1, so we have all kinds of things that > could be going wrong. > > The USB boots with the error: > > F1 FreeBSD > F6 PXE > Boot: F1 > > Not ufs > Not ufs > No /boot/loader > > When I investigated the flash's disklabel, I saw: > > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > a: 4000106 16unused0 0 > c: 40001220unused0 0 # "raw" part, don't > edit > > I newfs'd this flash drive and put data on it. The obvious thing to > do is set the disklabel for partition a to 4.2BSD. Even if I have to > reinstall the data, no big deal. > > So, run bsdlabel -e /dev/da0s1 and change the disklabel to: > > #size offsetfstype [fsize bsize bps/cpg] > a: 4000106 164.2BSD0 0 > c: 40001220unused 0 0 # "raw" part, don't > edit > > Then I get the error: > > bsdlabel: Class not found > re-edit the label? [y]: > > The same error appears if I try to edit the disklabel off-disk and > restore (-R). > > Google tells me that this error is not uncommon, but the only > solutions I saw (loading geom_mbr and setting kern.geom.debugflags=16) > did not help. > > Anyone have any suggestions? Or, can anyone tell me I'm running down > the wrong path? > > Thanks, > ==ml > > -- > Michael W. Lucasmwlu...@blackhelicopters.org, mwlu...@freebsd.org > http://www.BlackHelicopters.org/~mwlucas/<http://www.BlackHelicopters.org/%7Emwlucas/> > Latest book: Cisco Routers for the Desperate, 2nd Edition > http://www.CiscoRoutersForTheDesperate.com/ > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to " > freebsd-questions-unsubscr...@freebsd.org" > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
"disklabel: Class not found" when editing USB disk
Hi, I'm following Ceri's instructions for getting FreeBSD on a flash drive. My laptop is a Feb 1 -current, I'm installing FreeBSD 7.1, and the instructions are for 6.1, so we have all kinds of things that could be going wrong. The USB boots with the error: F1 FreeBSD F6 PXE Boot: F1 Not ufs Not ufs No /boot/loader When I investigated the flash's disklabel, I saw: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 4000106 16unused0 0 c: 40001220unused0 0 # "raw" part, don't edit I newfs'd this flash drive and put data on it. The obvious thing to do is set the disklabel for partition a to 4.2BSD. Even if I have to reinstall the data, no big deal. So, run bsdlabel -e /dev/da0s1 and change the disklabel to: #size offsetfstype [fsize bsize bps/cpg] a: 4000106 164.2BSD0 0 c: 40001220unused0 0 # "raw" part, don't edit Then I get the error: bsdlabel: Class not found re-edit the label? [y]: The same error appears if I try to edit the disklabel off-disk and restore (-R). Google tells me that this error is not uncommon, but the only solutions I saw (loading geom_mbr and setting kern.geom.debugflags=16) did not help. Anyone have any suggestions? Or, can anyone tell me I'm running down the wrong path? Thanks, ==ml -- Michael W. Lucasmwlu...@blackhelicopters.org, mwlu...@freebsd.org http://www.BlackHelicopters.org/~mwlucas/ Latest book: Cisco Routers for the Desperate, 2nd Edition http://www.CiscoRoutersForTheDesperate.com/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
disklabel not returning values for large RAID
Hi all- I recently added a 4th 750GB disk to my existing 1.5TB 3ware RAID5 volume, in an attempt to bring my total capacity up to ~2.25TB (4x750GB RAID5). I successfully used the 3ware 'tw-cli' utility to perform "Online Capacity Expansion." The controller migrated the existing data across all 4 disks, and it now reports a total volume capacity of 2095.44GB. Excellent. So, the issue I'm having is with getting FreeBSD 7.0-Release to recognize the additional space. I've seen scant few articles on the subject, the best of which is here: http://ezine.daemonnews.org/200111/growfs.html My primary issue right now is that when I attempt to run disklabel on the volume (/dev/da0) in order to get my calculations for fdisk, i get: # disklabel /dev/da0 disklabel: disks with more than 2^32-1 sectors are not supported and when I attempt it on the slice, i get: # disklabel /dev/da0s1 disklabel: /dev/da0s1: no valid label found What's the trick here? Thanks in advance, Darren David ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: disklabel not returning values for large RAID
Darren David wrote: Hi all- I recently added a 4th 750GB disk to my existing 1.5TB 3ware RAID5 volume, in an attempt to bring my total capacity up to ~2.25TB (4x750GB RAID5). I successfully used the 3ware 'tw-cli' utility to perform "Online Capacity Expansion." The controller migrated the existing data across all 4 disks, and it now reports a total volume capacity of 2095.44GB. Excellent. So, the issue I'm having is with getting FreeBSD 7.0-Release to recognize the additional space. I've seen scant few articles on the subject, the best of which is here: http://ezine.daemonnews.org/200111/growfs.html My primary issue right now is that when I attempt to run disklabel on the volume (/dev/da0) in order to get my calculations for fdisk, i get: # disklabel /dev/da0 disklabel: disks with more than 2^32-1 sectors are not supported and when I attempt it on the slice, i get: # disklabel /dev/da0s1 disklabel: /dev/da0s1: no valid label found What's the trick here? Thanks in advance, Darren David Update: OK, I discovered gpt. Here's the output: # gpt show /dev/da0 startsize index contents 0 1 MBR 1 62 63 2929629402 1 MBR part 165 2929629465 1464835815 Can anyone please give me some insight as to where to go from here? Cheers, Darren ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: need help with disklabel, "expected rawoffset 0, found 32"
Jerry McAllister wrote: On Tue, Oct 21, 2008 at 09:53:52AM +0200, Per olof Ljungmark wrote: Hi, I think sysinstall? got it wrong here and I get the complaint in the subject line on boot. This is amd64 if that matters. Nothing edited by hand. I must admit I don't fully understand what is going on here, "found 32" but the offsets are 63... I do not know what is causing this, but I think the offset of the 'c' partition (and the first real partition (a in this case)) should be 0 I have seen this a couple of times a long time ago and don't remember what happened other than I think I just arbitrarily set those offsets to 0 and it worked. Can you try booting up the fixit shell and hitting the disk with a manual fdisk and bsdlabel to see what happens. Also, you might try doing the dd(1) thing dd if=/dev/zero of=/dev/amrd0 bs=512 count=1000 before the fdisk and then another one after creating the slices manually dd if=/dev/zero of=/dev/amrd0s1 bs=512 count=1000 Note, that count value is arbitrary. That should clean up any junk on the drive. Also, I haven't seen/dealt with a disk device called amrd0 before. It appears to be something from one of the raid setups? So, maybe doing the dd thing might mangle that although, once it is a device, it should work the same as a drive. I ain't rich enough to have one of those raids to play with, though, so if someone else says otherwise, believe them. This is a backup server that uses da0 as a spool disk, we recreated that one so it looks ok. However, we still see "expected rawoffset 0, found 32" when booting and shutdown. It appears to me that sysinstall rewrote the MBR for amrd0 even if to my best knowledge it was not touched by us. Below is some information in case someone with insight might see the issue. Please let me know what more I can provide, thanks! camcontrol devlist -v scbus0 on ciss0 bus 0: at scbus0 target 0 lun 0 (pass0,da0) scbus1 on ciss0 bus 32: scbus2 on mpt0 bus 0: at scbus2 target 0 lun 0 (pass1,ch0) at scbus2 target 5 lun 0 (sa0,pass2) < > at scbus2 target -1 lun -1 () scbus3 on mpt1 bus 0: at scbus3 target 2 lun 0 (sa1,pass3) at scbus3 target 3 lun 0 (sa2,pass4) < > at scbus3 target -1 lun -1 () scbus4 on amr0 bus 0: at scbus4 target 0 lun 0 (pass5) at scbus4 target 1 lun 0 (pass6) at scbus4 target 2 lun 0 (pass7) at scbus4 target 3 lun 0 (pass8) at scbus4 target 4 lun 0 (pass9) at scbus4 target 5 lun 0 (pass10) scbus-1 on xpt0 bus 0: < > at scbus-1 target -1 lun -1 (xpt0) disklabel -A amrd0s1: # /dev/amrd0s1: type: ESDI disk: amrd0s1 label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 65535 sectors/unit: 2929674240 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 419430404.2BSD 2048 16384 28528 b: 8388608 4194304 swap c: 29296615320unused0 0 # "raw" part, don't edit d: 31457280 125829124.2BSD 2048 16384 28528 e: 2097152 440401924.2BSD 2048 16384 28528 f: 41943040 461373444.2BSD 2048 16384 28528 g: 2841581148 88080384 4.2BSD 2048 16384 28528 disklabel: partition c doesn't cover the whole unit! disklabel: An incorrect partition c may cause problems for standard system utilities disklabel -A da0s1 # /dev/da0s1: type: unknown disk: amnesiac label: flags: bytes/sector: 512 sectors/track: 32 tracks/cylinder: 255 sectors/cylinder: 8160 cylinders: 105414 sectors/unit: 860184297 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 8601842970unused0 0 # "raw" part, don't edit d: 860184233 634.2BSD 2048 16384 28528 fdisk -tv *** Working on device /dev/amrd0 *** parameters extracted from in-core disklabel are: cylinders=182363 heads=255 sectors/track=63 (16065 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=182363 heads=255 sectors/track=63 (16065 blks/cyl) Media sector size is 512 Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 1 is: sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) start 63, size 2929661532 (1430498 Meg), flag 80 (active) beg: cyl 0/ head 1/ sector 1; end: cyl 1023/
Re: need help with disklabel, "expected rawoffset 0, found 32"
On Tue, Oct 21, 2008 at 09:53:52AM +0200, Per olof Ljungmark wrote: > Hi, > > I think sysinstall? got it wrong here and I get the complaint in the > subject line on boot. This is amd64 if that matters. Nothing edited by hand. > > I must admit I don't fully understand what is going on here, "found 32" > but the offsets are 63... > > Filesystem on LSI controller amr(4): > > # /dev/amrd0s1a: > type: ESDI > disk: amrd0s1 > label: > flags: > bytes/sector: 512 > sectors/track: 63 > tracks/cylinder: 255 > sectors/cylinder: 16065 > cylinders: 65535 > sectors/unit: 2929674240 > rpm: 3600 > interleave: 1 > trackskew: 0 > cylinderskew: 0 > headswitch: 0 # milliseconds > track-to-track seek: 0 # milliseconds > drivedata: 0 > > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > a: 4194304 634.2BSD 2048 16384 28528 > b: 8388608 4194367 swap > c: 2929661532 63unused0 0 # "raw" part, > don't edit > d: 31457280 125829754.2BSD 2048 16384 28528 > e: 2097152 440402554.2BSD 2048 16384 28528 > f: 41943040 461374074.2BSD 2048 16384 28528 > g: 2841581148 880804474.2BSD 2048 16384 28528 > bsdlabel: partition c doesn't start at 0! > bsdlabel: partition c doesn't cover the whole unit! > bsdlabel: An incorrect partition c may cause problems for standard > system utilities I do not know what is causing this, but I think the offset of the 'c' partition (and the first real partition (a in this case)) should be 0 I have seen this a couple of times a long time ago and don't remember what happened other than I think I just arbitrarily set those offsets to 0 and it worked. Can you try booting up the fixit shell and hitting the disk with a manual fdisk and bsdlabel to see what happens. Also, you might try doing the dd(1) thing dd if=/dev/zero of=/dev/amrd0 bs=512 count=1000 before the fdisk and then another one after creating the slices manually dd if=/dev/zero of=/dev/amrd0s1 bs=512 count=1000 Note, that count value is arbitrary. That should clean up any junk on the drive. Also, I haven't seen/dealt with a disk device called amrd0 before. It appears to be something from one of the raid setups? So, maybe doing the dd thing might mangle that although, once it is a device, it should work the same as a drive. I ain't rich enough to have one of those raids to play with, though, so if someone else says otherwise, believe them. jerry > > > Filesystem on SmartArray controller ciss(4): > > # /dev/da0s1d: > type: SCSI > disk: da0s1 > label: > flags: > bytes/sector: 512 > sectors/track: 63 > tracks/cylinder: 255 > sectors/cylinder: 16065 > cylinders: 53544 > sectors/unit: 860192344 > rpm: 3600 > interleave: 1 > trackskew: 0 > cylinderskew: 0 > headswitch: 0 # milliseconds > track-to-track seek: 0 # milliseconds > drivedata: 0 > > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > c: 860184297 63unused0 0 # "raw" part, > don't edit > d: 860184297 634.2BSD 2048 16384 28528 > bsdlabel: partition c doesn't start at 0! > bsdlabel: partition c doesn't cover the whole unit! > bsdlabel: An incorrect partition c may cause problems for standard > system utilities > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: need help with disklabel, "expected rawoffset 0, found 32"
Per olof Ljungmark wrote: > I must admit I don't fully understand what is going on here ... The c partition should cover exactly the slice. For example, my ad0s1 is like that: The data for partition 1 is: sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) start 63, size 81915372 (39997 Meg), flag 80 (active) Now let us look at the label on this slice: lilas# disklabel ad0s1 # /dev/ad0s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 104857604.2BSD 2048 16384 8 b: 4126240 1048576 swap c: 819153720unused0 0 # "raw" part, don't edit d: 4159488 51748164.2BSD 2048 16384 28552 e: 72581068 93343044.2BSD 2048 16384 28552 You can see that the c) partition starts at offset 0 and has exactly the size 81915372 reported above. In your case you start at offset 63. Note that the first partition a) should start at offset 16 (see the "offset" entry in man bsdlabel) but this is not enforced in sysinstall. -- Michel TALON ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
need help with disklabel, "expected rawoffset 0, found 32"
Hi, I think sysinstall? got it wrong here and I get the complaint in the subject line on boot. This is amd64 if that matters. Nothing edited by hand. I must admit I don't fully understand what is going on here, "found 32" but the offsets are 63... Filesystem on LSI controller amr(4): # /dev/amrd0s1a: type: ESDI disk: amrd0s1 label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 65535 sectors/unit: 2929674240 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 4194304 634.2BSD 2048 16384 28528 b: 8388608 4194367 swap c: 2929661532 63unused0 0 # "raw" part, don't edit d: 31457280 125829754.2BSD 2048 16384 28528 e: 2097152 440402554.2BSD 2048 16384 28528 f: 41943040 461374074.2BSD 2048 16384 28528 g: 2841581148 880804474.2BSD 2048 16384 28528 bsdlabel: partition c doesn't start at 0! bsdlabel: partition c doesn't cover the whole unit! bsdlabel: An incorrect partition c may cause problems for standard system utilities Filesystem on SmartArray controller ciss(4): # /dev/da0s1d: type: SCSI disk: da0s1 label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 53544 sectors/unit: 860192344 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 860184297 63unused0 0 # "raw" part, don't edit d: 860184297 634.2BSD 2048 16384 28528 bsdlabel: partition c doesn't start at 0! bsdlabel: partition c doesn't cover the whole unit! bsdlabel: An incorrect partition c may cause problems for standard system utilities ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel error
Good day all, I am trying to install 7.0 RC1 on an old Dell laptop. However it fails to create the swap partition. The message I keep getting is: "Unable to find device node for /dev/ad0s1b in /dev!. The creation of filesystem will be aborted." Is there a way around it? Thanks in advance, Michael Michael Sherman http://msherman77.blogspot.com/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel does not write disklabel
bsdlabel: partition c doesn't cover the whole unit! bsdlabel: An incorrect partition c may cause problems for standard system probably because of that.. other reason - the device is open by other process ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel does not write disklabel
Hello guys! I tried to recover my partitions, which I remeved by accident. I used scan_ffs to create a new disklabel, which found all partitions, but when I use disklabel with -e or -R it does not write the table down. I think there's something I forgot or did wrong perhaps, but I'm not getting what it is. I put the log on http://home.schottelius.org/~nico/unix/freebsd/bsdlabel-edit-notsave Perhaps someone has an idea what I'm doing wrong. Sincerly Nico -- Think about Free and Open Source Software (FOSS). http://nico.schottelius.org/documentations/foss/the-term-foss/ PGP: BFE4 C736 ABE5 406F 8F42 F7CF B8BE F92A 9885 188C signature.asc Description: Digital signature
Re: Disklabel, partition d is usable or not ?
On Wed, Jul 18, 2007 at 12:44:02PM +0200, Gabriel Linder wrote: > Hi, > > This may sound as a dumb question, but during my 6.2-RELEASE (i386) setup I > notice the following in the handbook : > > Remember [...] that partitions b, c, and d have conventional meanings > that you should adhere to. > > But the partition d is used by sysinstall (with both automatic defaults and > manual setup), maybe this entry should be fixed if d has no more special > meaning ? I think you're right. The 'b' partition is usually used for swap space, while the 'c' partition represents the whole disk. The 'd' partition can be used normally, AFAIK. Additionally, only the 'a' partition can be booted from, IIRC. Roland -- R.F.Smith http://www.xs4all.nl/~rsmith/ [plain text _non-HTML_ PGP/GnuPG encrypted/signed email much appreciated] pgp: 1A2B 477F 9970 BA3C 2914 B7CE 1277 EFB0 C321 A725 (KeyID: C321A725) pgptQUAQi7KXD.pgp Description: PGP signature
Disklabel, partition d is usable or not ?
Hi, This may sound as a dumb question, but during my 6.2-RELEASE (i386) setup I notice the following in the handbook : > Remember [...] that partitions b, c, and d have conventional meanings that you should adhere to. But the partition d is used by sysinstall (with both automatic defaults and manual setup), maybe this entry should be fixed if d has no more special meaning ? Please note that this is the first time I use FreeBSD and this mailing-list :-) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel error in 4.11 release
On Tue, Apr 10, 2007 at 01:21:18PM -0700, fbsd bsd wrote: > Hey *, having a bit of a problem with disklabel. A bit of background. > > I'm building a lab with Olive boxes in it > (http://juniper.cluepon.net/index.php/Olive), which are basically i386 > machines running JunOS. Please note that this whole procedure is entirely > unsupported by Juniper and I would never *ever* recommend anyone run > production traffic on an Olive box. At any rate, it requires a base install > of FBSD 4.11, which you then do a pkg_add of their software over. > > The problem I'm having is that the package add fails every time. I > personally have no experience with disklabel so it took me a bit to track > down, but it seems like the raw device is not allowing anything to edit the > first sectors. Keep in mind that an Olive install effectively turns your BSD > box into a Juniper box, so the boot procedure will be different, which is why > I'm leaning toward this. Also, this is the output of disklabel querying the > kernel, then the device. > > > olive1# disklabel ad4 > # /dev/ad4: > type: unknown > disk: amnesiac > label: fictitious > flags: > bytes/sector: 512 > sectors/track: 63 > tracks/cylinder: 255 > sectors/cylinder: 16065 > cylinders: 9726 > sectors/unit: 15625 > rpm: 3600 > interleave: 1 > trackskew: 0 > cylinderskew: 0 > headswitch: 0 # milliseconds > track-to-track seek: 0 # milliseconds > drivedata: 0 > > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > c: 15625 0unused 0 0# (Cyl.0 - 9726*) > olive1# > olive1# > olive1# > olive1# > olive1# > olive1# > olive1# disklabel -r ad4 > disklabel: bad pack magic number (label is damaged, or pack is unlabeled) > olive1# > > > Attempts to use disklabel -e to edit the device label end in: > :q > disklabel: Operation not supported by device > re-edit the label? [y]: n > olive1# I don't know ab out the pkg_add below but that seems to be a later problem anyway. First comes disklabel issues. Three things to consider: Do you really have a drive 'ad4'? that would be the 5th ide type drive on the machine. If not, then you need to straighten out your drive designators. If you do have an ad4 (see dmesg to check), then you are attempting what they call a 'dangerously dedicated' drive by putting on a label without creating slices and putting the label in there. Maybe that is what is expected if this system you are trying to install mucks with the boot sector. But, it could also indicate you are not quite doing what you want. Be sure about what you are doing. What drive/slice/partition are you booted to? If it is to ad4 then the system probably will not allow you (or pkg_add) to write to that disk slice table or partition table - which, by the way, must be done as root. > > Finally, here is the failure of the pkg_add: > olive1# pkg_add jinstall-7.2R4.2-domestic-signed.tgz > Verified SHA1 checksum of jinstall-7.2R4.2-domestic.tgz > Adding jinstall... I don't see anywhere that you are saying where to install this stuff so I assume it is on the drive that you are running from. In that case, is it trying to write to a drive slice table or boot block that you are running from?As above, that isn't allowed. Anyway, as is probably obvious, I am not familiar with this piece of software you are trying to install so I don't know just what to expect from it. One more thing to ask: Is it really necessary to try to install it in such an old version of FreeBSD? Will it not work in FreeBSD 6.2? That would be better if it can be done. jerry > sysctl: unknown oid 'hw.re.model' > disklabel: bad pack magic number (label is damaged, or pack is unlabeled) > > WARNING: This package will load JUNOS 7.2R4.2 software. > WARNING: It will save JUNOS configuration files, and SSH keys > WARNING: (if configured), but erase all other files and information > WARNING: stored on this machine. It will attempt to preserve dumps > WARNING: and log files, but this can not be guaranteed. This is the > WARNING: pre-installation stage and all the software is loaded when > WARNING: you reboot the system. > > Saving the config files ... > Installing the bootstrap installer ... > disklabel: bad pack magic number (label is damaged, or pack is unlabeled) > > WARNING: Failed while trying to install bootstrap loaders > > Deleting bootstrap installer ... > disklabel: bad pack magic number (label is damaged, or pack is unlabeled) > > WARNING: This installation attempt will be aborted. > WARNING: If you wish to force the installation despite these warni
Re: disklabel error in 4.11 release
At 03:21 PM 4/10/2007, fbsd bsd wrote: Hey *, having a bit of a problem with disklabel. A bit of background. I'm building a lab with Olive boxes in it (http://juniper.cluepon.net/index.php/Olive), which are basically i386 machines running JunOS. Please note that this whole procedure is entirely unsupported by Juniper and I would never *ever* recommend anyone run production traffic on an Olive box. At any rate, it requires a base install of FBSD 4.11, which you then do a pkg_add of their software over. The problem I'm having is that the package add fails every time. I personally have no experience with disklabel so it took me a bit to track down, but it seems like the raw device is not allowing anything to edit the first sectors. Keep in mind that an Olive install effectively turns your BSD box into a Juniper box, so the boot procedure will be different, which is why I'm leaning toward this. Also, this is the output of disklabel querying the kernel, then the device. olive1# disklabel ad4 # /dev/ad4: type: unknown disk: amnesiac label: fictitious flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 9726 sectors/unit: 15625 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 156250unused0 0# (Cyl.0 - 9726*) olive1# olive1# olive1# olive1# olive1# olive1# olive1# disklabel -r ad4 disklabel: bad pack magic number (label is damaged, or pack is unlabeled) olive1# Attempts to use disklabel -e to edit the device label end in: :q disklabel: Operation not supported by device re-edit the label? [y]: n olive1# Finally, here is the failure of the pkg_add: olive1# pkg_add jinstall-7.2R4.2-domestic-signed.tgz Verified SHA1 checksum of jinstall-7.2R4.2-domestic.tgz Adding jinstall... sysctl: unknown oid 'hw.re.model' disklabel: bad pack magic number (label is damaged, or pack is unlabeled) WARNING: This package will load JUNOS 7.2R4.2 software. WARNING: It will save JUNOS configuration files, and SSH keys WARNING: (if configured), but erase all other files and information WARNING: stored on this machine. It will attempt to preserve dumps WARNING: and log files, but this can not be guaranteed. This is the WARNING: pre-installation stage and all the software is loaded when WARNING: you reboot the system. Saving the config files ... Installing the bootstrap installer ... disklabel: bad pack magic number (label is damaged, or pack is unlabeled) WARNING: Failed while trying to install bootstrap loaders Deleting bootstrap installer ... disklabel: bad pack magic number (label is damaged, or pack is unlabeled) WARNING: This installation attempt will be aborted. WARNING: If you wish to force the installation despite these warnings WARNING: you may use the 'force' option on the command line. pkg_add: install script returned error status pkg_add: install script returned error status olive1# If anyone has any tips I'd appreciate it. sysinstall will label the disks as part of the install. Is the hardware preventing the boot area from being written to? It would appear that this is the case. -Derek -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. MailScanner thanks transtec Computers for their support. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel error in 4.11 release
Hey *, having a bit of a problem with disklabel. A bit of background. I'm building a lab with Olive boxes in it (http://juniper.cluepon.net/index.php/Olive), which are basically i386 machines running JunOS. Please note that this whole procedure is entirely unsupported by Juniper and I would never *ever* recommend anyone run production traffic on an Olive box. At any rate, it requires a base install of FBSD 4.11, which you then do a pkg_add of their software over. The problem I'm having is that the package add fails every time. I personally have no experience with disklabel so it took me a bit to track down, but it seems like the raw device is not allowing anything to edit the first sectors. Keep in mind that an Olive install effectively turns your BSD box into a Juniper box, so the boot procedure will be different, which is why I'm leaning toward this. Also, this is the output of disklabel querying the kernel, then the device. olive1# disklabel ad4 # /dev/ad4: type: unknown disk: amnesiac label: fictitious flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 9726 sectors/unit: 15625 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 156250unused0 0# (Cyl.0 - 9726*) olive1# olive1# olive1# olive1# olive1# olive1# olive1# disklabel -r ad4 disklabel: bad pack magic number (label is damaged, or pack is unlabeled) olive1# Attempts to use disklabel -e to edit the device label end in: :q disklabel: Operation not supported by device re-edit the label? [y]: n olive1# Finally, here is the failure of the pkg_add: olive1# pkg_add jinstall-7.2R4.2-domestic-signed.tgz Verified SHA1 checksum of jinstall-7.2R4.2-domestic.tgz Adding jinstall... sysctl: unknown oid 'hw.re.model' disklabel: bad pack magic number (label is damaged, or pack is unlabeled) WARNING: This package will load JUNOS 7.2R4.2 software. WARNING: It will save JUNOS configuration files, and SSH keys WARNING: (if configured), but erase all other files and information WARNING: stored on this machine. It will attempt to preserve dumps WARNING: and log files, but this can not be guaranteed. This is the WARNING: pre-installation stage and all the software is loaded when WARNING: you reboot the system. Saving the config files ... Installing the bootstrap installer ... disklabel: bad pack magic number (label is damaged, or pack is unlabeled) WARNING: Failed while trying to install bootstrap loaders Deleting bootstrap installer ... disklabel: bad pack magic number (label is damaged, or pack is unlabeled) WARNING: This installation attempt will be aborted. WARNING: If you wish to force the installation despite these warnings WARNING: you may use the 'force' option on the command line. pkg_add: install script returned error status pkg_add: install script returned error status olive1# If anyone has any tips I'd appreciate it. - Be a PS3 game guru. Get your game face on with the latest PS3 news and previews at Yahoo! Games. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Recovering bsdlabel / disklabel with scan_ffs
This is not a question, just for the archives if someone encountered a similar problem. Perhaps there's an easier way to recover a lost bsdlabel / disklabel though... While trying to rip a DVD with sysutils/vobcopy on 6.2-RC1, the system suddenly froze and could not reboot anymore. Not even the boot loader would come up after this. After swapping disks (putting a brand new FreeBSD 6.2-RC1 HDD as primary and the previous disk as secondary), only /dev/ad3s1 slice would appear, but no more /dev/ad3s1a, /dev/ad3s1d, ... partitions. Running # fdisk /dev/ad3 showed inconsistant (overlapping etc...) results for all slices as well, instead of the usual output of a fully dedicated disk, which should have looked like this: sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) start 63, size 156360582 (76347 Meg), flag 80 (active) beg: cyl 0/ head 1/ sector 1; end: cyl 1023/ head 10/ sector 63 The data for partition 2 is: The data for partition 3 is: The data for partition 4 is: Obviously, the boot sector has been badly damanged. After restoring the partition table (allocating whole disk to FreeBSD-slice), and adding the BootMgr using /usr/sbin/sysinstall, # bsdlabel /dev/ad3s1 still didn't show the old partitions. Uh-oh. Bad news: no backups, no backup or printout of bsdlabel; and I didn't exactly remember the size and layout of the partitions on that machine. Enters /usr/ports/sysutils/scan_ffs. Calling: # scan_ffs /dev/ad3s1 showed lines like these: ufs2 at 0 size 262144 mount / time Sat Apr 10 01:08:46 2004 ufs2 at 5242880 size 4194304 mount /usr time Sat Apr 10 01:08:57 2004 ... Wonderful! There's a catch here: while the offsets (at ...) are the ones to add when editing the bsdlabel (bsdlabel -e /dev/ad3s1), the sizes aren't (the partitions wouldn't fsck -n). In fact, I had to use (size*4) here, e.g.: # /dev/ad0s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 104857604.2BSD 2048 16384 8 b: 4194304 1048576 swap c: 1563605820unused0 0 # "raw" part, don't edit d: 16777216 52428804.2BSD 2048 16384 28552 e: 16777216 220200964.2BSD 2048 16384 28552 [...] Obviously, this had something to do with fsize being 2048 by newfs defaults (and not 512): size*2048 bytes blocks = (size*4)*512 bytes blocks Second catch: If you can't remember the SLICE coordinates, you could run scan_ffs on the raw disk with: # scan_ffs /dev/ad3 instead of # scan_ffs /dev/ad3s1 but all offsets would be off-by-(offset-of-the-slice), e.g.: ufs2 at 63 size 262144 mount / time Sat Apr 10 01:08:46 2004 (note: 63 instead of 0; 63 was start offset of the FreeBSD slice). There's another catch: GBDE encrypted partitions can't (for obvious reasons) be detected with scan_ffs. As long as you don't have two contiguous GBDE partitions, it's possible to infer offset and size from the surrounding partitions (I was lucky enough to have such a friendly layout on this machine: one GBDE partitions in the middle of the slice, and another one at the end). Fortunately, and thanks to scan_ffs and some head-scratching, I was able to restore the whole system (and all user-data), with one notable exception: fsck choked and quit on the filesystem holding /usr/local with a message like: cannot alloc 553234321 bytes for inostathead Mounting that filesystem read-only showed that there were no valuable data in there that couldn't be recreated by newfs and recompiling all ports. To summarize: scan_ffs is a real life saver, but: * Don't put two (or more) encrypted partitions side-by-side * Remember to scale the size output of scan_ffs (I had to x 4) * Infer missing information (size/offset of swap and encrypted partitions) from surrounding partitions if possible. * Back up the output of: # fdisk /dev/ad0 (and other disks) # bsdlabel /dev/ad0s1 (and other FreeBSD slices) and GBDE/GEOM keys somewhere else. * Don't be lazy backing up valuable data... ;-) scan_ffs is such an incredibly useful emergency tool, it should really be part of the fixit and freesbie CDs... ;) Good luck! Regards, -cpghost. -- Cordula's Web. http://www.cordula.ws/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel and usb device
On Mon, Dec 11, 2006 at 09:26:31PM -0500, Peter Matulis wrote: > I am having trouble viewing my USB compact flash reader with my FBSD > 5.5 system. I have done so in the past. For some reason I can no > longer do so. > > This is what I'm getting: > > # disklabel /dev/da0s1 > disklabel: /dev/da0s1: no valid label found That is the message I get when I forget to 'su' before trying to run disklabel/bsdlabel. By the way, I don't think there is a disklabel in FreeBSD 5.5. It is bsdlabel, but I suppose that is just a typing thing in your message above. Below you mention the card has an OpenBSD file system. I have never used that, so don't know if it makes any difference. jerry > > # fdisk /dev/da0s1 > *** Working on device /dev/da0s1 *** > parameters extracted from in-core disklabel are: > cylinders=249 heads=64 sectors/track=32 (2048 blks/cyl) > > parameters to be used for BIOS calculations are: > cylinders=249 heads=64 sectors/track=32 (2048 blks/cyl) > > Media sector size is 512 > Warning: BIOS sector numbering starts with sector 1 > Information from DOS bootblock is: > The data for partition 1 is: > sysid 108 (0x6c),(unknown) > start 1684955424, size 1701998624 (831054 Meg), flag a > beg: cyl 368/ head 82/ sector 37; > end: cyl 357/ head 97/ sector 35 > The data for partition 2 is: > sysid 110 (0x6e),(unknown) > start 1998616933, size 544105832 (265676 Meg), flag 73 > beg: cyl 97/ head 115/ sector 32; > end: cyl 107/ head 121/ sector 32 > The data for partition 3 is: > sysid 121 (0x79),(QNX4.x 3rd part) > start 538988361, size 538976288 (263172 Meg), flag 72 > beg: cyl 356/ head 101/ sector 33; > end: cyl 0/ head 13/ sector 10 > The data for partition 4 is: > sysid 83 (0x53),(DM6 Aux3) > start 1394614304, size 21337 (10 Meg), flag 53 > beg: cyl 333/ head 89/ sector 19; > end: cyl 339/ head 68/ sector 15 > > > >From logs: > > kernel: umass0: SanDisk ImageMate 8 in 1, rev 2.00/91.44, addr 2 > > > The card contains an OpenBSD filesystem. > > Peter > > __ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel and usb device
I am having trouble viewing my USB compact flash reader with my FBSD 5.5 system. I have done so in the past. For some reason I can no longer do so. This is what I'm getting: # disklabel /dev/da0s1 disklabel: /dev/da0s1: no valid label found # fdisk /dev/da0s1 *** Working on device /dev/da0s1 *** parameters extracted from in-core disklabel are: cylinders=249 heads=64 sectors/track=32 (2048 blks/cyl) parameters to be used for BIOS calculations are: cylinders=249 heads=64 sectors/track=32 (2048 blks/cyl) Media sector size is 512 Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 1 is: sysid 108 (0x6c),(unknown) start 1684955424, size 1701998624 (831054 Meg), flag a beg: cyl 368/ head 82/ sector 37; end: cyl 357/ head 97/ sector 35 The data for partition 2 is: sysid 110 (0x6e),(unknown) start 1998616933, size 544105832 (265676 Meg), flag 73 beg: cyl 97/ head 115/ sector 32; end: cyl 107/ head 121/ sector 32 The data for partition 3 is: sysid 121 (0x79),(QNX4.x 3rd part) start 538988361, size 538976288 (263172 Meg), flag 72 beg: cyl 356/ head 101/ sector 33; end: cyl 0/ head 13/ sector 10 The data for partition 4 is: sysid 83 (0x53),(DM6 Aux3) start 1394614304, size 21337 (10 Meg), flag 53 beg: cyl 333/ head 89/ sector 19; end: cyl 339/ head 68/ sector 15 >From logs: kernel: umass0: SanDisk ImageMate 8 in 1, rev 2.00/91.44, addr 2 The card contains an OpenBSD filesystem. Peter __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Questions re: disklabel for external USB drives
On Mon, Dec 11, 2006 at 09:34:59PM +0100, Joerg Pernfuss wrote: > On Mon, 11 Dec 2006 14:17:29 -0500 > Jerry McAllister <[EMAIL PROTECTED]> wrote: > > > > Is there an easy way to force the device to work as something like: > > > > > > /dev/da1s1d > > > > > > on all of the servers, including ones that do not already have a > > > SCSI disk subsystem and existing /dev/da0 devices? > > > > I think, in FreeBSD SCSI device stuff, you can force it to be > > something, but I have never done it and don't know how - and since > > it doesn't matter, don't see the reason to try. > > A imho better solution is to load the geom_label class. > Then either give the device itself a label: > da3s1d -> label/usbstick42s1d > or set the UFS Label field of the filesystems via 'tunefs -L', > this will give you for example > da5s1a -> ufs/ustick5data > da5s1d -> ufs/ustick5keys > > or whatever you set the label to, obviously. If geom_label is loaded, > you have unique device names on all servers (if you don't mix things > up when you label them). OK. But, that really sounds like more work than just making your fstab file relevant to the machine it is on and not worrying about the rest. jerry > > Regards, > Joerg > > -- > | /"\ ASCII ribbon | GnuPG Key ID | e86d b753 3deb e749 6c3a | > | \ / campaign against |0xbbcaad24 | 5706 1f7d 6cfd bbca ad24 | > | XHTML in email |.the next sentence is true. | > | / \ and news | .the previous sentence was a lie.| ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Questions re: disklabel for external USB drives
On Mon, 11 Dec 2006 14:17:29 -0500 Jerry McAllister <[EMAIL PROTECTED]> wrote: > > Is there an easy way to force the device to work as something like: > > > > /dev/da1s1d > > > > on all of the servers, including ones that do not already have a > > SCSI disk subsystem and existing /dev/da0 devices? > > I think, in FreeBSD SCSI device stuff, you can force it to be > something, but I have never done it and don't know how - and since > it doesn't matter, don't see the reason to try. A imho better solution is to load the geom_label class. Then either give the device itself a label: da3s1d -> label/usbstick42s1d or set the UFS Label field of the filesystems via 'tunefs -L', this will give you for example da5s1a -> ufs/ustick5data da5s1d -> ufs/ustick5keys or whatever you set the label to, obviously. If geom_label is loaded, you have unique device names on all servers (if you don't mix things up when you label them). Regards, Joerg -- | /"\ ASCII ribbon | GnuPG Key ID | e86d b753 3deb e749 6c3a | | \ / campaign against |0xbbcaad24 | 5706 1f7d 6cfd bbca ad24 | | XHTML in email |.the next sentence is true. | | / \ and news | .the previous sentence was a lie.| signature.asc Description: PGP signature
Re: Questions re: disklabel for external USB drives
On Mon, Dec 11, 2006 at 02:26:42PM -0500, [EMAIL PROTECTED] wrote: > On Mon, 11 Dec 2006, Jerry McAllister wrote: > > > On Mon, Dec 11, 2006 at 02:05:27PM -0500, [EMAIL PROTECTED] wrote: > > > > > > > > I just got an external USB drive that I want to use for disk-based > > > backups. It is important that this drive be useable on different FreeBSD > > > servers that we have. > > > > > > I got it working on a test server ok, but I noticed that the sysinstall > > > utility labeled the device as: > > > > > > /dev/da0s1d > > > > > > Since the test server only has an IDE drive, that's fine, but this > > > external USB drive needs to be able to work on productions servers that > > > already have SCSI and SAS devices, one of which already uses that label > > > for its active "/usr" partition. > > > > > > Is there an easy way to force the device to work as something like: > > > > > > /dev/da1s1d > > > > > > on all of the servers, including ones that do not already have a SCSI disk > > > subsystem and existing /dev/da0 devices? > > > > You don't have to do that unless you are worried about getting confused. > > If you put the drive on a machine that already has da0 used up, it will > > magically become da1. The label doesn't have anything to do with > > whether it is da0 or da1. That is determined by its position on the > > controller. > > > > I think, in FreeBSD SCSI device stuff, you can force it to be > > something, but I have never done it and don't know how - and since > > it doesn't matter, don't see the reason to try. > > > > jerry > > Cool! Thanks! Of course, you have to keep track of the different device labels when you mount the file systems and/or put them in /etc/fstab on whichever machine so they mount the right device for that machine. jerry > > James Smallacombe PlantageNet, Inc. CEO and Janitor > [EMAIL PROTECTED] > http://3.am > = > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Questions re: disklabel for external USB drives
On Mon, Dec 11, 2006 at 02:05:27PM -0500, [EMAIL PROTECTED] wrote: > > I just got an external USB drive that I want to use for disk-based > backups. It is important that this drive be useable on different FreeBSD > servers that we have. > > I got it working on a test server ok, but I noticed that the sysinstall > utility labeled the device as: > > /dev/da0s1d > > Since the test server only has an IDE drive, that's fine, but this > external USB drive needs to be able to work on productions servers that > already have SCSI and SAS devices, one of which already uses that label > for its active "/usr" partition. > > Is there an easy way to force the device to work as something like: > > /dev/da1s1d > > on all of the servers, including ones that do not already have a SCSI disk > subsystem and existing /dev/da0 devices? You don't have to do that unless you are worried about getting confused. If you put the drive on a machine that already has da0 used up, it will magically become da1. The label doesn't have anything to do with whether it is da0 or da1. That is determined by its position on the controller. I think, in FreeBSD SCSI device stuff, you can force it to be something, but I have never done it and don't know how - and since it doesn't matter, don't see the reason to try. jerry > > TIA, > > James Smallacombe PlantageNet, Inc. CEO and Janitor > [EMAIL PROTECTED] > http://3.am > = > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Questions re: disklabel for external USB drives
I just got an external USB drive that I want to use for disk-based backups. It is important that this drive be useable on different FreeBSD servers that we have. I got it working on a test server ok, but I noticed that the sysinstall utility labeled the device as: /dev/da0s1d Since the test server only has an IDE drive, that's fine, but this external USB drive needs to be able to work on productions servers that already have SCSI and SAS devices, one of which already uses that label for its active "/usr" partition. Is there an easy way to force the device to work as something like: /dev/da1s1d on all of the servers, including ones that do not already have a SCSI disk subsystem and existing /dev/da0 devices? TIA, James Smallacombe PlantageNet, Inc. CEO and Janitor [EMAIL PROTECTED] http://3.am = ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: FreeBSD 6.x and disklabel
On 10/30/06 12:05, Reuben A. Popp wrote: Good morning everyone, Recently, we've been looking at purchasing a SAN here and I came across this site while doing some research. Seeing as how we just met with reps from Apple to discuss their offerings, I thought that the article was well worth reading ;) http://www.mostlygeek.com/node/39 Anyway, I realize that the article deals with the 5.x branch, so it may not be 100% exact when it comes to our implementation (6.x). The article makes note that as of 5.x, there were many parts that were still 32 bit, which in turn affected the maximum filesystem size (~2TB). If you haven't already gotten an answer The following may help: http://www.freebsd.org/projects/bigdisk/index.html Not that we would need a filesystem larger than that, but does anyone know if this is still an issue, or was it changed in 6.x, or if there are plans to rework it in -CURRENT? TIA :) Reuben A. Popp -- Regards, Eric ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
FreeBSD 6.x and disklabel
Good morning everyone, Recently, we've been looking at purchasing a SAN here and I came across this site while doing some research. Seeing as how we just met with reps from Apple to discuss their offerings, I thought that the article was well worth reading ;) http://www.mostlygeek.com/node/39 Anyway, I realize that the article deals with the 5.x branch, so it may not be 100% exact when it comes to our implementation (6.x). The article makes note that as of 5.x, there were many parts that were still 32 bit, which in turn affected the maximum filesystem size (~2TB). Not that we would need a filesystem larger than that, but does anyone know if this is still an issue, or was it changed in 6.x, or if there are plans to rework it in -CURRENT? TIA :) Reuben A. Popp -- Reuben A. Popp Interim Systems Administrator Information Technology Department East Central College 1+ 636 583-5195 ext 2480 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel question
On Sat, Oct 21, 2006 at 12:33:17AM +0800, James Villa wrote: > # /dev/ad6s1: > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > a: 122880004.2BSD 2048 16384 11272 > b: 2097152 1228800 swap > c: 1023982470unused0 0 # "raw" part, don't > edi > t > d: 1228800 33259524.2BSD 2048 16384 11272 > e: 1228800 45547524.2BSD 2048 16384 11272 > f: 4096 57835524.2BSD 2048 16384 28552 > g: 55654695 467435524.2BSD 2048 16384 28552 > > > as single user i run: disklabel -e /dev/ad6s1 and i add: > > h: 53903178 0 4.2BSD 2048 16384 28552 > is this correct? Wouldn't this overlap/wipe out partitions 1, b, d, e, f and part of g? What are you trying to do here? It looks like your disk slice is all used up. There is no room to add an 'h' partition without deleting some of the existing stuff. jerry > from sysinstall: > unuse > size(ST): 53903178 > end: 156301487 > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel question
# /dev/ad6s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 122880004.2BSD 2048 16384 11272 b: 2097152 1228800 swap c: 1023982470unused0 0 # "raw" part, don't edi t d: 1228800 33259524.2BSD 2048 16384 11272 e: 1228800 45547524.2BSD 2048 16384 11272 f: 4096 57835524.2BSD 2048 16384 28552 g: 55654695 467435524.2BSD 2048 16384 28552 as single user i run: disklabel -e /dev/ad6s1 and i add: h: 53903178 0 4.2BSD 2048 16384 28552 is this correct? from sysinstall: unuse size(ST): 53903178 end: 156301487 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Adjusting partition size with disklabel
On 6/30/06, Morten A. Middelthon <[EMAIL PROTECTED]> wrote: Hi, long story short, I have a partition on a RAID5 array which after an accident where I had to rebuild the array became smaller than it originally was. Here's the original size: amrd1: 1430505MB (2929674240 sectors) RAID 5 (degraded) and the new size after the rebuild: amrd1: 1430400MB (2929459200 sectors) RAID 5 (optimal) Is it possible to use 'bsdlabel -e' to shrink the partition down to a size which will fit the new size of the array? To my knowledge, you can only growfs(8) them, not shrink them. References: http://www.freebsd.org/cgi/man.cgi?query=growfs&apropos=0&sektion=0&manpath=FreeBSD+6.1-RELEASE&format=html -- Joao Barros ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Adjusting partition size with disklabel
Hi, long story short, I have a partition on a RAID5 array which after an accident where I had to rebuild the array became smaller than it originally was. Here's the original size: amrd1: 1430505MB (2929674240 sectors) RAID 5 (degraded) and the new size after the rebuild: amrd1: 1430400MB (2929459200 sectors) RAID 5 (optimal) Is it possible to use 'bsdlabel -e' to shrink the partition down to a size which will fit the new size of the array? with regards, -- Morten A. Middelthon Which is worse: ignorance or apathy? Who knows? Who cares? pgpxrrQIOAHyR.pgp Description: PGP signature
Creating a disklabel for NetBSD slice
Hi, I have FreeBSD 6.1 and NetBSD 3.0 on my machine. I can make disklabel entries (in NetBSD) for the FreeBSD partitions, and that way mount them in NetBSD. Just a matter of giving the absolute offset values of the partitions. But I cant find any straight forward way of mounting NetBSD partitions under FreeBSD. Doing "disklabel /dev/ad0s2" (my NetBSD slice) gives an error message that there's no valid label to be found. So I make up a disklabel for ad0s2. I get the NetBSD disklabel into a file, edit it to make the number of partitions less than 8, remove all the miscellaneous info, change all the offsets to relative values, and then make a disklabel thus: "disklabel -R ad0s2 nbsd.txt" ("nbsd.txt" being the file which contains the disklabels). After this the disklabel is created fine, but when I boot into NetBSD, the disklabel there is messed up and so NetBSD can't load. I had a backup of the disklabels anyways (was expecting something like this), so I managed to get it fixed. Booted into a NetBSD install CD and restored the disklabel. And now when I boot into FreeBSD I see that its lost whatever disklabel I had written. So my question is this: is there any way I can get FreeBSD to create a disklabel for ad0s2, but *not overwrite* the NetBSD one? I mean, I see frequent references to "on-disk" label and "in-core" label in the manpage, and I was wondering maybe its possible to create a disklabel that's internal to FreeBSD and doesn't really overwrite the NetBSD one. Is that possible? What are these "in-core" and "on-disk" labels anyways? Thanks, Rakhesh -- NetBSD/i386 3.0 + pkgsrc-current | OpenBSD/i386 3.9 Toshiba Satellite L10-102 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel messup.
On 3/11/06, Steve P. <[EMAIL PROTECTED]> wrote: > I have been reading up on how to "clone" a disk, so I can boot into a copy. > Seems like some say tar can do it, but I have seen a place that said to only > use dump. I tend to use pax, since it's a bit more straightforward* than tar for copying, but dump/restore is the one true way, if you have an investment in the validity of the copy. tar (and pax) require that you make the slice bootable (via bsdlabel or else- wise), the dump and restore cycle cares not a whit for such conventions. *lies, pax is about as backwards and arcane as you could possibly hope for. -- -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel messup.
Illoai, You were right. There is nothing wrong. Part of my problem was that I had never had or seen anything on other slices. Now I have two fbsd installs, ad0s1 and ad02 and I mistook ad0s2's missing info as applying to ad0s1. What I really want to do is use ad0s1 as my production install and ad0s2 as my test install. I guess my procedure would be to copy prod to test, then perform any testing for new ports, etc. Then once tested, perform those same actions on ad0s1. I have been reading up on how to "clone" a disk, so I can boot into a copy. Seems like some say tar can do it, but I have seen a place that said to only use dump. Oh, well. I am not quite there yet, working on baby steps. - Original Message - From: "[EMAIL PROTECTED]" To: "Steve P." Subject: Re: disklabel messup. Date: Sat, 11 Mar 2006 04:33:25 -0600 On 3/10/06, Steve P. wrote: > I was using sysinstall's disklabel facility to poke around. I > accidentally did "Undo" on my installed 6.0 working slice. Are you sure you actually did anything? If /etc/fstab shows them correctly still and running 'df' shows them still as they should be, I'm pretty sure that running sysinstall again will show the slices and partitions correctly. I would advise that 'bsdlabel /dev/ad0' will show you your label just as effectively, and that df(1) is probably the correct method of gazing in rapture at your mounted partitions. Other commands to read up on might be fdisk and diskinfo. Good luck. -- -- -- ___ Play 100s of games for FREE! http://games.mail.com/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel messup.
On 3/10/06, Steve P. <[EMAIL PROTECTED]> wrote: > I was using sysinstall's disklabel facility to poke around. I > accidentally did "Undo" on my installed 6.0 working slice. Are you sure you actually did anything? If /etc/fstab shows them correctly still and running 'df' shows them still as they should be, I'm pretty sure that running sysinstall again will show the slices and partitions correctly. I would advise that 'bsdlabel /dev/ad0' will show you your label just as effectively, and that df(1) is probably the correct method of gazing in rapture at your mounted partitions. Other commands to read up on might be fdisk and diskinfo. Good luck. -- -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel messup.
I was using sysinstall's disklabel facility to poke around. I accidentally did "Undo" on my installed 6.0 working slice. Now, the mount points for my partitions fail to appear, even though I did not "write" them. I just exited sysinstall. The odd thing is that the system shows no adverse effects even on shutdown reboot. I guess /etc/fstab provides the info to reboot. What is the best way to get the mount points, etc back so sysinstall/disklabel shows them correctly? Would I use the disklabel command directly to reconstruct? Thanks. Steve. -- ___ Play 100s of games for FREE! http://games.mail.com/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel output
On 2006-03-03 10:02, Grant Peel <[EMAIL PROTECTED]> wrote: > This drive is a SCSI 74 GIG drive. > > The machine has been freezing on me lately, out of the blue, > with no log errors enetered. No crash dumps created. The only > fix is to cold boot. > > Here is the disklabel output Should I be worried? > > root on s1# disklabel /dev/da0s1 > # /dev/da0s1: > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > a: 2097152 20971524.2BSD 2048 16384 28552 > b: 20971520 swap > c: 1433639970unused0 0 # "raw" part, don't > edit > d: 4194304 41943044.2BSD 2048 16384 28552 > e: 12582912 83886084.2BSD 2048 16384 28552 > f: 122392477 209715204.2BSD 2048 16384 28552 > root on s1# > > root on s1# disklabel /dev/da0s1a > disklabel: /dev/da0s1a: no valid label found The /dev/da0s1a device *is* already part of a disk label. You shouldn't normally install a label or use any 'label' found on the start of such a partition. > root on s1# disklabel /dev/da0s1b > root on s1# disklabel /dev/da0s1c > root on s1# disklabel /dev/da0s1d Similarly, looking at a 'label' here doesn't help at all. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel output
This drive is a SCSI 74 GIG drive. The machine has been freezing on me lately, out of the blue, with no log errors enetered. No crash dumps created. The only fix is to cold boot. Here is the disklabel output Should I be worried? root on s1# disklabel /dev/da0s1 # /dev/da0s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 2097152 20971524.2BSD 2048 16384 28552 b: 20971520 swap c: 1433639970unused0 0 # "raw" part, don't edit d: 4194304 41943044.2BSD 2048 16384 28552 e: 12582912 83886084.2BSD 2048 16384 28552 f: 122392477 209715204.2BSD 2048 16384 28552 root on s1# root on s1# disklabel /dev/da0s1a disklabel: /dev/da0s1a: no valid label found root on s1# disklabel /dev/da0s1b # /dev/da0s1b: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 2097152 20972154.2BSD 2048 16384 28552 b: 2097152 63 swap c: 143363997 63unused0 0 # "raw" part, don't edit d: 4194304 41943674.2BSD 2048 16384 28552 e: 12582912 83886714.2BSD 2048 16384 28552 f: 122392477 209715834.2BSD 2048 16384 28552 partition a: offset past end of unit partition a: partition extends past end of unit partition b: partition extends past end of unit partition c: partition extends past end of unit disklabel: partition c doesn't start at 0! disklabel: partition c doesn't cover the whole unit! disklabel: An incorrect partition c may cause problems for standard system utilities partition d: offset past end of unit partition d: partition extends past end of unit partition e: offset past end of unit partition e: partition extends past end of unit partition f: offset past end of unit partition f: partition extends past end of unit root on s1# disklabel /dev/da0s1c # /dev/da0s1c: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 2097152 20972154.2BSD 2048 16384 28552 b: 2097152 63 swap c: 143363997 63unused0 0 # "raw" part, don't edit d: 4194304 41943674.2BSD 2048 16384 28552 e: 12582912 83886714.2BSD 2048 16384 28552 f: 122392477 209715834.2BSD 2048 16384 28552 partition c: partition extends past end of unit disklabel: partition c doesn't start at 0! disklabel: An incorrect partition c may cause problems for standard system utilities partition f: partition extends past end of unit root on s1# disklabel /dev/da0s1d disklabel: /dev/da0s1d: no valid label found ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Can't mount existing data disk after upgrade (disklabel gone)
On 1/20/06, Joe S <[EMAIL PROTECTED]> wrote: > First of all, I have learned something VERY important about FreeBSD. > SAVE your DISKLABELS! > When I try to mount /dev/ad4s1a to /data, I get this error: > coruscant# mount /dev/ad4s1a /mnt > mount: /dev/ad4s1a on /mnt: incorrect super block > > How can I properly re-create the disklabel? It seems this is what I need > to do in order to get to my data. Any help would be greatly appreciated. Silly question: did you: newfs -U /dev/ad4s1a ? -- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
RE: Can't mount existing data disk after upgrade (disklabel gone)
Joe, If you did not touch ad4 the disklabel it should still have been there, but I presume that you have destroyed it by now. Did you make any kind of backup ? such as a dump of the filesystem ? How did you try to mount ad4s1a ? ( I am assuming you tried to mount the 'partition' and not the slice (ad4s1) or the device (ad4)). Why are you changing the offset of ad4s1a from 16 to 2097215 ? Is there another partition on there somewhere or do you just not want to use that part of your disk ? Also, you scan_ffs ad4 instead of ad4s1 which is what you should be interested in. The size differences are normal because there's a difference between ad4 ad4s1 and ad4s1a. You want to write the correct disklabel to ad4s1 and then mount ad4s1a. You're 'a' partition extends past the size of the disk (c) because a(584002180)+offset(2097215)= 586099395 > c:( 586099332) (by 63) If you can correctly replicate the exact disklabel you should be able to access your data. Otherwise, chalk it up to experience. I hope this helps. Ruben -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Joe S Sent: January 20, 2006 6:18 PM To: freebsd-questions@freebsd.org Cc: Joe S Subject: Can't mount existing data disk after upgrade (disklabel gone) First of all, I have learned something VERY important about FreeBSD. SAVE your DISKLABELS! I have 2 drives in my home file server. One 80GB drive (ad0) for the OS and one 300 GB drive for my data. A few days ago, I performed a clean install of FreeBSD 6.0 on the 1st drive (ad0). After the install I tried to mount ad4, but could not. I read somewhere that I was supposed to save my disklabels. DOH! I did not do this, nor have I ever read about this in the FreeBSD handbook. After reading through the handbook and googling, I found a tool called scan_ffs that can help me recreate my disklabels by scanning my drive for partitions. Great! Here is the output of scan_ffs on ad4: [EMAIL PROTECTED] scan_ffs -l /dev/ad4 X: 584002180 2097215 4.2BSD 2048 16384 0 # /data Good. That is my /data partition. Its UFS2, created in FBSD 5.4. Next, I tried editing the disklabel. It starts out like this: # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099316 16unused0 0 c: 5860993320unused0 0 I changed it to this: # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 584002180 2097215 4.2BSD 2048 16384 0 # /data c: 5860993320unused0 0 But, I get this error: partition a: partition extends past end of unit re-edit the label? [y]: Here is the output of bsdlabel. Notice the different sizes: [EMAIL PROTECTED] bsdlabel -r ad4 # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586114688 16unused0 0 c: 5861147040unused0 0 [EMAIL PROTECTED] bsdlabel -r ad4s1 # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099316 16unused0 0 c: 5860993320unused0 0 [EMAIL PROTECTED] [EMAIL PROTECTED] bsdlabel -r ad4s1a # /dev/ad4s1a: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099300 16unused0 0 c: 5860993160unused0 0 When I try to mount /dev/ad4s1a to /data, I get this error: coruscant# mount /dev/ad4s1a /mnt mount: /dev/ad4s1a on /mnt: incorrect super block How can I properly re-create the disklabel? It seems this is what I need to do in order to get to my data. Any help would be greatly appreciated. -joe PS> Here is my dmesg: Copyright (c) 1992-2005 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 6.0-SECURITY #0: Mon Dec 19 23:46:33 UTC 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC ACPI APIC Table: Timecounter "i8254" frequency 1193182 Hz quality 0 CPU: Intel(R) Pentium(R) 4 CPU 2.40GHz (2394.01-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0xf29 Stepping = 9 Features=0xbfebfbff Features2=0x4400> real memory = 1072889856 (1023 MB) avail memory = 1040945152 (992 MB) ioapic0 irqs 0-23 on motherboard npx0: [FAST] npx0: on motherboard npx0: INT 16 interface acpi0: on motherboard acpi0: Power Button (fixed) pci_link0: irq 11 on acpi0 pci_link1: irq 3 on acpi0 pci_link2: irq 10 on acpi0 pci_link3: irq 5 on acpi0 pci_link4: irq 0 on acpi0 pci_link5: irq 0 on acpi0 pci_link6: irq 9 on acpi0 pci_link7: irq 9 on acpi0 Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000 acpi_timer0: <24-bit timer at 3.579545MHz> port 0x408-0x40b on acpi0 cpu0: on acpi0 acpi_
Re: Can't mount existing data disk after upgrade (disklabel gone)
Joe S wrote: First of all, I have learned something VERY important about FreeBSD. SAVE your DISKLABELS! I have 2 drives in my home file server. One 80GB drive (ad0) for the OS and one 300 GB drive for my data. A few days ago, I performed a clean install of FreeBSD 6.0 on the 1st drive (ad0). After the install I tried to mount ad4, but could not. I read somewhere that I was supposed to save my disklabels. DOH! I did not do this, nor have I ever read about this in the FreeBSD handbook. After reading through the handbook and googling, I found a tool called scan_ffs that can help me recreate my disklabels by scanning my drive for partitions. Great! Here is the output of scan_ffs on ad4: [EMAIL PROTECTED] scan_ffs -l /dev/ad4 X: 584002180 2097215 4.2BSD 2048 16384 0 # /data Good. That is my /data partition. Its UFS2, created in FBSD 5.4. Next, I tried editing the disklabel. It starts out like this: # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099316 16unused0 0 c: 5860993320unused0 0 I changed it to this: # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 584002180 2097215 4.2BSD 2048 16384 0 # /data c: 5860993320unused0 0 But, I get this error: partition a: partition extends past end of unit re-edit the label? [y]: Here is the output of bsdlabel. Notice the different sizes: [EMAIL PROTECTED] bsdlabel -r ad4 # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586114688 16unused0 0 c: 5861147040unused0 0 [EMAIL PROTECTED] bsdlabel -r ad4s1 # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099316 16unused0 0 c: 5860993320unused0 0 [EMAIL PROTECTED] [EMAIL PROTECTED] bsdlabel -r ad4s1a # /dev/ad4s1a: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099300 16unused0 0 c: 5860993160unused0 0 When I try to mount /dev/ad4s1a to /data, I get this error: coruscant# mount /dev/ad4s1a /mnt mount: /dev/ad4s1a on /mnt: incorrect super block How can I properly re-create the disklabel? It seems this is what I need to do in order to get to my data. Any help would be greatly appreciated. -joe PS> Here is my dmesg: Copyright (c) 1992-2005 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 6.0-SECURITY #0: Mon Dec 19 23:46:33 UTC 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC ACPI APIC Table: Timecounter "i8254" frequency 1193182 Hz quality 0 CPU: Intel(R) Pentium(R) 4 CPU 2.40GHz (2394.01-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0xf29 Stepping = 9 Features=0xbfebfbff Features2=0x4400> real memory = 1072889856 (1023 MB) avail memory = 1040945152 (992 MB) ioapic0 irqs 0-23 on motherboard npx0: [FAST] npx0: on motherboard npx0: INT 16 interface acpi0: on motherboard acpi0: Power Button (fixed) pci_link0: irq 11 on acpi0 pci_link1: irq 3 on acpi0 pci_link2: irq 10 on acpi0 pci_link3: irq 5 on acpi0 pci_link4: irq 0 on acpi0 pci_link5: irq 0 on acpi0 pci_link6: irq 9 on acpi0 pci_link7: irq 9 on acpi0 Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000 acpi_timer0: <24-bit timer at 3.579545MHz> port 0x408-0x40b on acpi0 cpu0: on acpi0 acpi_throttle0: on cpu0 pcib0: port 0xcf8-0xcff on acpi0 pci0: on pcib0 agp0: mem 0xf800-0xfbff at device 0.0 on pci0 pcib1: at device 1.0 on pci0 pci1: on pcib1 pci1: at device 0.0 (no driver attached) uhci0: port 0xcc00-0xcc1f irq 16 at device 29.0 on pci0 uhci0: [GIANT-LOCKED] usb0: on uhci0 usb0: USB revision 1.0 uhub0: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub0: 2 ports with 2 removable, self powered uhci1: port 0xd000-0xd01f irq 19 at device 29.1 on pci0 uhci1: [GIANT-LOCKED] usb1: on uhci1 usb1: USB revision 1.0 uhub1: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub1: 2 ports with 2 removable, self powered uhci2: port 0xd400-0xd41f irq 18 at device 29.2 on pci0 uhci2: [GIANT-LOCKED] usb2: on uhci2 usb2: USB revision 1.0 uhub2: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub2: 2 ports with 2 removable, self powered uhci3: port 0xd800-0xd81f irq 16 at device 29.3 on pci0 uhci3: [GIANT-LOCKED] usb3: on uhci3 usb3: USB revision 1.0 uhub3: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub3: 2 ports with 2 removable, self powered ehci0: mem 0xfebffc00-0xfebf irq 23 at device 29.7 on pci0 ehci0: [GIANT-LOCKED] usb4: EHCI version 1.0 usb4: companion controllers, 2 ports each: usb0 usb1 usb2 usb3 usb4: on ehci0 usb4: USB revision 2.0 uhub4
Can't mount existing data disk after upgrade (disklabel gone)
First of all, I have learned something VERY important about FreeBSD. SAVE your DISKLABELS! I have 2 drives in my home file server. One 80GB drive (ad0) for the OS and one 300 GB drive for my data. A few days ago, I performed a clean install of FreeBSD 6.0 on the 1st drive (ad0). After the install I tried to mount ad4, but could not. I read somewhere that I was supposed to save my disklabels. DOH! I did not do this, nor have I ever read about this in the FreeBSD handbook. After reading through the handbook and googling, I found a tool called scan_ffs that can help me recreate my disklabels by scanning my drive for partitions. Great! Here is the output of scan_ffs on ad4: [EMAIL PROTECTED] scan_ffs -l /dev/ad4 X: 584002180 2097215 4.2BSD 2048 16384 0 # /data Good. That is my /data partition. Its UFS2, created in FBSD 5.4. Next, I tried editing the disklabel. It starts out like this: # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099316 16unused0 0 c: 5860993320unused0 0 I changed it to this: # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 584002180 2097215 4.2BSD 2048 16384 0 # /data c: 5860993320unused0 0 But, I get this error: partition a: partition extends past end of unit re-edit the label? [y]: Here is the output of bsdlabel. Notice the different sizes: [EMAIL PROTECTED] bsdlabel -r ad4 # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586114688 16unused0 0 c: 5861147040unused0 0 [EMAIL PROTECTED] bsdlabel -r ad4s1 # /dev/ad4s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099316 16unused0 0 c: 5860993320unused0 0 [EMAIL PROTECTED] [EMAIL PROTECTED] bsdlabel -r ad4s1a # /dev/ad4s1a: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 586099300 16unused0 0 c: 5860993160unused0 0 When I try to mount /dev/ad4s1a to /data, I get this error: coruscant# mount /dev/ad4s1a /mnt mount: /dev/ad4s1a on /mnt: incorrect super block How can I properly re-create the disklabel? It seems this is what I need to do in order to get to my data. Any help would be greatly appreciated. -joe PS> Here is my dmesg: Copyright (c) 1992-2005 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 6.0-SECURITY #0: Mon Dec 19 23:46:33 UTC 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC ACPI APIC Table: Timecounter "i8254" frequency 1193182 Hz quality 0 CPU: Intel(R) Pentium(R) 4 CPU 2.40GHz (2394.01-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0xf29 Stepping = 9 Features=0xbfebfbff Features2=0x4400> real memory = 1072889856 (1023 MB) avail memory = 1040945152 (992 MB) ioapic0 irqs 0-23 on motherboard npx0: [FAST] npx0: on motherboard npx0: INT 16 interface acpi0: on motherboard acpi0: Power Button (fixed) pci_link0: irq 11 on acpi0 pci_link1: irq 3 on acpi0 pci_link2: irq 10 on acpi0 pci_link3: irq 5 on acpi0 pci_link4: irq 0 on acpi0 pci_link5: irq 0 on acpi0 pci_link6: irq 9 on acpi0 pci_link7: irq 9 on acpi0 Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000 acpi_timer0: <24-bit timer at 3.579545MHz> port 0x408-0x40b on acpi0 cpu0: on acpi0 acpi_throttle0: on cpu0 pcib0: port 0xcf8-0xcff on acpi0 pci0: on pcib0 agp0: mem 0xf800-0xfbff at device 0.0 on pci0 pcib1: at device 1.0 on pci0 pci1: on pcib1 pci1: at device 0.0 (no driver attached) uhci0: port 0xcc00-0xcc1f irq 16 at device 29.0 on pci0 uhci0: [GIANT-LOCKED] usb0: on uhci0 usb0: USB revision 1.0 uhub0: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub0: 2 ports with 2 removable, self powered uhci1: port 0xd000-0xd01f irq 19 at device 29.1 on pci0 uhci1: [GIANT-LOCKED] usb1: on uhci1 usb1: USB revision 1.0 uhub1: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub1: 2 ports with 2 removable, self powered uhci2: port 0xd400-0xd41f irq 18 at device 29.2 on pci0 uhci2: [GIANT-LOCKED] usb2: on uhci2 usb2: USB revision 1.0 uhub2: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub2: 2 ports with 2 removable, self powered uhci3: port 0xd800-0xd81f irq 16 at device 29.3 on pci0 uhci3: [GIANT-LOCKED] usb3: on uhci3 usb3: USB revision 1.0 uhub3: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub3: 2 ports with 2 removable, self powered ehci0: mem 0xfebffc00-0xfebf irq 23 at device 29.7 on pci0 ehci0: [GIANT-LOCKED] usb4: EHCI version 1.0 usb4: companion controllers, 2 ports each: usb0 usb1 usb2 usb3 usb4: on ehci0 usb4: USB revision 2.0 uhub4: Intel EHCI root h
Re: Help deriving a corrupted disklabel
Hi Anish, Thanks for the reply -- scan_ffs did exactly what I needed it to do! For those who encounter a similar problem (damaged disklabel on a FreeBSD slice), it's not as simple as one might hope, since I was limited to the "Fixit" shell on the FreeBSD 6.0-RELEASE CD. scan_ffs is not on the CD, which makes it much more difficult (configuring networking, obtaining it via ftp, unpacking the package, etc., since 'pkg_add -r' doesn't work in a Fixit shell). Setup was most of the work! Once I had a usable copy of scan_ffs, the rest was simple. Thanks again for the help. Without it, I'm sure I'd still be struggling to use dd and grep to locate magic numbers using a shell script. Doug On 12/2/05, Anish Mistry <[EMAIL PROTECTED]> wrote: > > On Wednesday 30 November 2005 09:15 pm, Doug H wrote: > > > > PROBLEM: bsdlabel showed me that slice 1 (FreeBSD 5.4) is damaged > > and only partition c existed and was incorrect. I do not have / > > cannot find a written copy of my disklabel for that disk (a good > > suggestion to *strongly emphasize* in the installation manual for > > newbies!). I did recall that ad1s1a ('/') was 512M, so I was able > > to write a label and mount that partition from a "Fixit" shell. > > > > QUESTIONS: How can I rederive the remaining disklabel for that > > disk? Could a copy possibly be stored somewhere on root if I > > didn't do it myself when building the system? > > > > > sysutils/scan_ffs > I've always used it from a emergency FreeBSD diagnostic CD (custom > Freesbie) and it works great. I've never been stuck with only a > fixit shell though. > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Help deriving a corrupted disklabel
On Wednesday 30 November 2005 09:15 pm, Doug H wrote: > One of my disks has 3 active partitions: FreeBSD 5.4-RC3, NTFS > (not-bootable), and FreeBSD 6.0-RELEASE. > > I developed problems while installing FreeBSD 6.0. Installation > went as well as can be expected using sysinstall (no difficulty > other than pulling packages from the CD), but when I rebooted, > nothing was bootable on that disk. I verified that the boot record > (using boot0) seems okay: slice table is fine (40G, 80G, 40G, and > 40G unused on 200G drive). > > Using bsdlabel, I confirmed that slice 3 (FreeBSD 6) is fine, but > for some reason I'm not concerned with now, is unbootable. > > PROBLEM: bsdlabel showed me that slice 1 (FreeBSD 5.4) is damaged > and only partition c existed and was incorrect. I do not have / > cannot find a written copy of my disklabel for that disk (a good > suggestion to *strongly emphasize* in the installation manual for > newbies!). I did recall that ad1s1a ('/') was 512M, so I was able > to write a label and mount that partition from a "Fixit" shell. > > QUESTIONS: How can I rederive the remaining disklabel for that > disk? Could a copy possibly be stored somewhere on root if I > didn't do it myself when building the system? > > I have not tried to boot from that root partition. Trying several > possible labels has resulted in "incorrect super block" errors for > the partitions after 'a'. Random guessing will be very tedious. > > My research has indicated that I could binary grep the raw ad1s1c > partition to locate the magic numbers for the super blocks and > derive the partitions from that information. I even found a little > 'c' language program Peter Dufault posted 11 years ago on this list > to locate magic numbers. > > My hope is that in 11 years of development, FreeBSD would have > created a clever tool to aid this process! I've found enough > entries in these lists to think that the effort would be justified > and much appreciated. If there is no tool, can someone tell me the > value of FS_UFS2_MAGIC? I presume that's what I should search for > - it's a UFS2 filesystem. Having only a "Fixit" shell is somewhat > limiting. > sysutils/scan_ffs I've always used it from a emergency FreeBSD diagnostic CD (custom Freesbie) and it works great. I've never been stuck with only a fixit shell though. -- Anish Mistry pgpEMb2pnEvII.pgp Description: PGP signature
Rebuilding a corrupt disklabel
I tried posting this 12 hours ago, but haven't seen it flow through the list... reposting with updates: One of my disks has 3 active partitions: FreeBSD 5.4-RC3, NTFS (not-bootable), and FreeBSD 6.0-RELEASE. I developed problems while installing FreeBSD 6.0. Installation went well, but when I rebooted, nothing was bootable on that disk. I verified that the boot record (using boot0) seems okay: slice table is fine (40G, 80G, 40G, and 40G unused on 200G drive). Using bsdlabel, I confirmed that slice 3 (FreeBSD 6) is fine, but for some reason I'm not concerned with now, is unbootable. PROBLEM: bsdlabel showed me that slice 1 (FreeBSD 5.4) is damaged and only partition c existed and was incorrect. I do not have & cannot find a written copy of my disklabel for that disk (a good suggestion to *strongly emphasize* in the installation manual for newbies!). I did recall that ad1s1a ('/') was 512M, so I was able to write a label and mount the root partition from a "Fixit" shell. QUESTIONS: How can I rederive the remaining disklabel for that disk? Could a copy possibly be stored somewhere on root if I didn't do it myself when building the system? I have not tried to boot from that root partition. Trying several possible labels has resulted in "incorrect super block" errors for the partitions after 'a'. Random guessing will be very tedious. My research has indicated that I could binary grep the raw ad1s1c partition to locate the magic numbers for the super blocks and derive the partitions from that information. I even found a little 'c' language program Peter Dufault posted 11 years ago on this list to locate magic numbers. My hope is that in 11 years of development, FreeBSD will have created a clever tool to aid this process! I've found enough entries while searching these lists to think that the effort would be justified and much appreciated. I found the value of FS_UFS2_MAGIC = 0x19540119. I presume that's what I should search for - it's a UFS2 filesystem. Having only a "Fixit" shell is somewhat limiting. I tried 'grep -ab -f pat /dev/ad1s1a' on a FreeBSD 5.1 installation, but it crashed with a swap error. The 'pat' file has the 4-byte magic number. It works with a small test file I created using vi. Thanks, Doug ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Help deriving a corrupted disklabel
One of my disks has 3 active partitions: FreeBSD 5.4-RC3, NTFS (not-bootable), and FreeBSD 6.0-RELEASE. I developed problems while installing FreeBSD 6.0. Installation went as well as can be expected using sysinstall (no difficulty other than pulling packages from the CD), but when I rebooted, nothing was bootable on that disk. I verified that the boot record (using boot0) seems okay: slice table is fine (40G, 80G, 40G, and 40G unused on 200G drive). Using bsdlabel, I confirmed that slice 3 (FreeBSD 6) is fine, but for some reason I'm not concerned with now, is unbootable. PROBLEM: bsdlabel showed me that slice 1 (FreeBSD 5.4) is damaged and only partition c existed and was incorrect. I do not have / cannot find a written copy of my disklabel for that disk (a good suggestion to *strongly emphasize* in the installation manual for newbies!). I did recall that ad1s1a ('/') was 512M, so I was able to write a label and mount that partition from a "Fixit" shell. QUESTIONS: How can I rederive the remaining disklabel for that disk? Could a copy possibly be stored somewhere on root if I didn't do it myself when building the system? I have not tried to boot from that root partition. Trying several possible labels has resulted in "incorrect super block" errors for the partitions after 'a'. Random guessing will be very tedious. My research has indicated that I could binary grep the raw ad1s1c partition to locate the magic numbers for the super blocks and derive the partitions from that information. I even found a little 'c' language program Peter Dufault posted 11 years ago on this list to locate magic numbers. My hope is that in 11 years of development, FreeBSD would have created a clever tool to aid this process! I've found enough entries in these lists to think that the effort would be justified and much appreciated. If there is no tool, can someone tell me the value of FS_UFS2_MAGIC? I presume that's what I should search for - it's a UFS2 filesystem. Having only a "Fixit" shell is somewhat limiting. Thanks, Doug ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Strange disklabel
The problem is resolved. I did a "dump, newfs, restore" and now the disklabel looks more reasonable: # disklabel da1s1 # /dev/da1s1: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 29271714570unused0 0 # "raw" part, don't edit d: 20971520004.2BSD 2048 16384 28552 e: 1048576000 2097152004.2BSD 2048 16384 28552 f: 1668880257 12582912004.2BSD 2048 16384 28552 I now consider this subject for closed. Anders Gytri --- On Thu, 4 Aug 2005, Anders Gytri wrote: > I have recently installed a big disk (1.4 TB sata raid with scsi > interface) on an i386 computer running FreeBSD 5.4-RELEASE-p1. I have 3 > partitions on the FreeBSD slice. The disk is working OK, but the disklabel > seem a little strange. For one of the partitions the "fsize bsize bps/cpg" > values are "0 0 0". For the other two partitions these values are > "2048 16384 28552" which I beleve is more normal. The filesystem is > initialized (with newfs). fsck gives no errors. > > Partial or full output from the commands "dmesg", "df", "mount", "fdisk > da1", "bsdlabel -A da1s1", "tunefs -p /dev/da1s1e", "fsck /dev/da1s1e" and > "dumpfs /dev/da1s1e" is attached below. > > Should I worry about this, or can I ignore it. > > Anders Gytri > - > # dmesg > da1 at ahd0 bus 0 target 0 lun 0 > da1: Fixed Direct Access SCSI-3 device > da1: 160.000MB/s transfers (80.000MHz, offset 127, 16bit), Tagged Queueing > Enabled > da1: 1429284MB (2927173632 512 byte sectors: 255H 63S/T 182208C) > - > # df > Filesystem 1K-blocks Used Avail Capacity Mounted on > /dev/da1s1e 507788836 47657476 41950825410%/home > - > # mount > /dev/da1s1e on /home (ufs, NFS exported, local, soft-updates) > - > # fdisk da1 > *** Working on device /dev/da1 *** > parameters extracted from in-core disklabel are: > cylinders=182208 heads=255 sectors/track=63 (16065 blks/cyl) > > Figures below won't work with BIOS for partitions not in cyl 1 > parameters to be used for BIOS calculations are: > cylinders=182208 heads=255 sectors/track=63 (16065 blks/cyl) > > Media sector size is 512 > Warning: BIOS sector numbering starts with sector 1 > Information from DOS bootblock is: > The data for partition 1 is: > sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) > start 63, size 2927171457 (1429282 Meg), flag 80 (active) > beg: cyl 0/ head 1/ sector 1; > end: cyl 959/ head 254/ sector 63 > The data for partition 2 is: > > The data for partition 3 is: > > The data for partition 4 is: > > - > # bsdlabel -A da1s1 > # /dev/da1s1: > type: unknown > disk: amnesiac > label: > flags: > bytes/sector: 512 > sectors/track: 63 > tracks/cylinder: 255 > sectors/cylinder: 16065 > cylinders: 182207 > sectors/unit: 2927171457 > rpm: 3600 > interleave: 1 > trackskew: 0 > cylinderskew: 0 > headswitch: 0 # milliseconds > track-to-track seek: 0 # milliseconds > drivedata: 0 > > 8 partitions: > #size offsetfstype [fsize bsize bps/cpg] > c: 29271714570unused0 0 # "raw" part, > don't edit > d: 20971520004.2BSD 2048 16384 28552 > e: 1048576000 2097152004.2BSD0 0 0 > f: 1668880257 12582912004.2BSD 2048 16384 28552 > - > # tunefs -p /dev/da1s1e > tunefs: ACLs: (-a) disabled > tunefs: MAC multilabel: (-l) disabled > tunefs: soft updates: (-n) enabled > tunefs: maximum blocks per file in a cylinder group: (-e) 2048 > tunefs: average file size: (-f)16384 > tunefs: average number of files in a directory: (-s) 64 > tunefs: minimum percentage of free space: (-m) 8% > tunefs: optimization preference: (-o) time > tunefs: volume label: (-L) > - > # fsck /dev/da1s1e > ** /dev/da1s1e > ** Last Mounted on /home > ** Phase 1 - Check Blocks and Sizes > ** Phase 2 - Check Pathnames > ** Phase 3 - Check Connectivity > ** Phase 4 - Check Reference Counts > ** Phase 5
Strange disklabel
I have recently installed a big disk (1.4 TB sata raid with scsi interface) on an i386 computer running FreeBSD 5.4-RELEASE-p1. I have 3 partitions on the FreeBSD slice. The disk is working OK, but the disklabel seem a little strange. For one of the partitions the "fsize bsize bps/cpg" values are "0 0 0". For the other two partitions these values are "2048 16384 28552" which I beleve is more normal. The filesystem is initialized (with newfs). fsck gives no errors. Partial or full output from the commands "dmesg", "df", "mount", "fdisk da1", "bsdlabel -A da1s1", "tunefs -p /dev/da1s1e", "fsck /dev/da1s1e" and "dumpfs /dev/da1s1e" is attached below. Should I worry about this, or can I ignore it. Anders Gytri - # dmesg da1 at ahd0 bus 0 target 0 lun 0 da1: Fixed Direct Access SCSI-3 device da1: 160.000MB/s transfers (80.000MHz, offset 127, 16bit), Tagged Queueing Enabled da1: 1429284MB (2927173632 512 byte sectors: 255H 63S/T 182208C) - # df Filesystem 1K-blocks Used Avail Capacity Mounted on /dev/da1s1e 507788836 47657476 41950825410%/home - # mount /dev/da1s1e on /home (ufs, NFS exported, local, soft-updates) ----- # fdisk da1 *** Working on device /dev/da1 *** parameters extracted from in-core disklabel are: cylinders=182208 heads=255 sectors/track=63 (16065 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=182208 heads=255 sectors/track=63 (16065 blks/cyl) Media sector size is 512 Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 1 is: sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) start 63, size 2927171457 (1429282 Meg), flag 80 (active) beg: cyl 0/ head 1/ sector 1; end: cyl 959/ head 254/ sector 63 The data for partition 2 is: The data for partition 3 is: The data for partition 4 is: - # bsdlabel -A da1s1 # /dev/da1s1: type: unknown disk: amnesiac label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 182207 sectors/unit: 2927171457 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 29271714570unused0 0 # "raw" part, don't edit d: 20971520004.2BSD 2048 16384 28552 e: 1048576000 2097152004.2BSD0 0 0 f: 1668880257 12582912004.2BSD 2048 16384 28552 - # tunefs -p /dev/da1s1e tunefs: ACLs: (-a) disabled tunefs: MAC multilabel: (-l) disabled tunefs: soft updates: (-n) enabled tunefs: maximum blocks per file in a cylinder group: (-e) 2048 tunefs: average file size: (-f)16384 tunefs: average number of files in a directory: (-s) 64 tunefs: minimum percentage of free space: (-m) 8% tunefs: optimization preference: (-o) time tunefs: volume label: (-L) - # fsck /dev/da1s1e ** /dev/da1s1e ** Last Mounted on /home ** Phase 1 - Check Blocks and Sizes ** Phase 2 - Check Pathnames ** Phase 3 - Check Connectivity ** Phase 4 - Check Reference Counts ** Phase 5 - Check Cyl groups 36339 files, 23828738 used, 230065680 free (712 frags, 28758121 blocks, 0.0% fragmentation) - # dumpfs /dev/da1s1e magic 19540119 (UFS2) timeThu Aug 4 10:32:28 2005 superblock location 65536 id [ 42a2003d 11af58d9 ] ncg 2787size262144000 blocks 253894418 bsize 16384 shift 14 mask0xc000 fsize 2048shift 11 mask0xf800 frag8 shift 3 fsbtodb 2 minfree 8% optim timesymlinklen 120 maxbsize 16384 maxbpg 2048maxcontig 8 contigsumsize 8 nbfree 28758121ndir423 nifree 65603083nffree 712 bpg 11761 fpg 94088 ipg 23552 nindir 2048inopb 64 maxfilesize 140806241583103 sbsize 2048cgsize 16384 csaddr 3000cssize 45056 sblkno 40 cblkno 48 iblkno 56 dblkno 3000 cgrotor 1939fmod0 ronly 0 clean 1 avgfpdir 64 avgfilesize 16384 flags soft-updates fsmnt /home volname swuid 0 ___ freebsd-quest
disklabel and boot2 on Freebsd 5.4
Hi, I was trying to put an alternative boot2 on a freebsd 5.4 box. With the -s option disappeared from 5.4 disklabel, how do I put a customized boot2 to a slice on FreeBSD 5.4? Regards, Ming ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Location of disklabel
On Thu, Mar 10, 2005 at 04:32:30PM -0500, Carl J wrote: > Hi all! To all your FS guru's outthere, I desperately need > to know where the disklabel is stored (since my disk is in trouble!) > > Situation: > > My /dev/ad0s1 has 2 partitions: "a" (FS) followed by "b" (swap). > By using "disklabel -r", I see my "a" and "b" indeed > take up the entire slice. > > My desperate question: > > Where, then, is the "disklabel" stored? The second sector of the slice that the disklabel is partitioning. For example, a disklabel on your first slice would be stored in the second sector of /dev/ad0s1. The command dd if=/dev/ad0s1 skip=1 | hexdump will give you a hexdump of the disklabel. Since the 'a' partition of the disklabel normally starts at the beginning of the slice that the disklabel is in, it is identical to reading from the slice directly, just a little shorter. Also, the 'c' partition always covers the entire slice so it is identical assuming the disklabel isn't messed up. > > Somewhere in the partition table? The Master Boot Record? > The reserved cylinder #0? No, msdos partition table that creates what are called slices in the bsd world reside in the last few byte of the Master boot record, but this has nothing to do with the disklabel that is stored in the slice. And normally the only thing you will find in cylinder 0 is the master boot record which is the very first sector of the hard disk. > > Or is it stored somewhere inside /dev/ad0s1a ?? > (if that's the case, does that mean the UFS1 > intentionally left some space unused, for this purpose? > And if so, is it always at a fixed location within a UFS1 slice?) Actually, since the 'a' partition is the same as the beginning of the slice it's in, the ufs filesystem always skips the first 16 sectors of whatever partition it's in. > > What if in my slice, I have SWAP first, and then UFS1, > then does that mean the SWAP Format also reserves > some unused space for the disklabel to go??? > > Sorry if the question is stupid. I just somehow couldn't > logically see where it would be stored, and yet be compatible > with having other OS on the same drive... etc. > > Thanks! > > - Carl > > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" -- I sense much NT in you. NT leads to Bluescreen. Bluescreen leads to downtime. Downtime leads to suffering. NT is the path to the darkside. Powerful Unix is. Public Key: ftp://ftp.tallye.com/pub/lorenl_pubkey.asc Fingerprint: CEE1 AAE2 F66C 59B5 34CA C415 6D35 E847 0118 A3D2 pgpSnGRsgOe1o.pgp Description: PGP signature
Location of disklabel
Hi all! To all your FS guru's outthere, I desperately need to know where the disklabel is stored (since my disk is in trouble!) Situation: My /dev/ad0s1 has 2 partitions: "a" (FS) followed by "b" (swap). By using "disklabel -r", I see my "a" and "b" indeed take up the entire slice. My desperate question: Where, then, is the "disklabel" stored? Somewhere in the partition table? The Master Boot Record? The reserved cylinder #0? Or is it stored somewhere inside /dev/ad0s1a ?? (if that's the case, does that mean the UFS1 intentionally left some space unused, for this purpose? And if so, is it always at a fixed location within a UFS1 slice?) What if in my slice, I have SWAP first, and then UFS1, then does that mean the SWAP Format also reserves some unused space for the disklabel to go??? Sorry if the question is stupid. I just somehow couldn't logically see where it would be stored, and yet be compatible with having other OS on the same drive... etc. Thanks! - Carl ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel disappeared after power loss
Thank you very much for the quick response! > > After a sudden power loss, one of the disks (ad2) can not > > recover. > What does `fdisk ad2' say? Some nonsense -- as if I had only a 30Mb partition-4... > What does `disklabel ad2' say? Something about "amnesiac" with only the c-partition. I used /stand/sysinstall to create a small swap partition at the beginning of the drive. I don't know, what it does, but it re-created the label, which I was then able to edit with disklabel. scan_ffs (from the sysutils/scan_ffs) helped me recover the exact size and offset. I wish, fsck had scan_ffs' functionality built-in... > Maybe something is messed up, so that disklabel does not dare to > write a new disklabel. Well, sysinstall did not mind... > Is something from ad2 mounted read-writeable, when you get the "Op > not perm" error? No, definetly not. > How about > 1. copying the data from the former ad2e into another filesystem, This is a 50% full 180Gb disk. The only other disk nearby is a 20Gb system drive... > 3. establishing an all new disklabel with proper ad2e? :-) > (most likely ad2e is too big?) ad2e was not too big -- it did not exist. But sysinstall did the job. Perhaps, disklabel needs to learn a few tricks from that tool. And, of course, the main question is, why could the label disappear as a result of something as mundane as powerloss? Yours, -mi ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel disappeared after power loss
Hello! This is a 4.11-STABLE from Dec 24. After a sudden power loss, one of the disks (ad2) can not recover. It was "dangerously dedicated" and had two partitions -- swap (ad2b) and data (ad2e). Any attempts to use either (swapon, fsck, mount) now result in EINVAL. `disklabel ad2' creates an imaginary label with only the ad2c covering the entire drive. If I try to add the ad2b and ad2e in disklabel (I remember the sizes), I get: disklabel: Operation not supported by device I can read from /dev/ad2 directly. How can I restore access to the filesystem? Thanks! -mi ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
CDROM Disc-1 install fails @ Xorg package choice | Disklabel Editor won't config 2nd RAID drive
Hi, I'm having some issues installing FreeBSD from the Disc-1 CDROM. I'm using a Dell SC420 PowerEdge with two 80GB drives in a RAID level-1 array. When I get to the Disklabel Editor, both drives are shown at top, and I can configure the first using auto defaults, but when I select the 2nd drive the display doesn't change - drive 1 is still shown in the Part column, no matter if I press "a" or anything. (side question: if and when it's all working, how will I know if the RAID is doing its thing; is there a utility for that?) Also when I get to the Xorg choices screen I get stuck in a loop - can't get on to the next config screen. Thanx, Jeff __ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel missing on a raid-5
Hi, I'm having a serious problem with one of my freebsd 4.6-R box. I have always used Freebsd to run my database servers and I've never had such a problem. If anyone knows the solution to my problem below, please let me know because I have some really important data that I badly need. Short story: Someone booted a database server by accident, it won't boot up and got hang up right after post (where the /|\ spinning supposedly begins) so that person booted it again, and again. (three times in total I think). Anyway, since it won't boot up at all, I've decided to install a IDE HD with Freebsd 4.6 on it, boot up the server and attempt to retrieve the database backups from the messed-up raid-5 HDDS that I stored in a /usr/home/backup folder. dmesg says: da0 at asr0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: Tagged Queueing Enabled da0: 70006MB (143372288 512 byte sectors: 255H 63S/T 8924C) Looks good. So I go #mount /dev/da0s1 /mnt It returns: Jan 28 23:22:31 /kernel: da0: cannot find label (no disk label) Jan 28 23:22:31 /kernel: da0s1: cannot find label (no disk label) And I am able to cd to mnt and read everything, but /usr and /var (that's where the data reside) are empty. Then I tried: #fdisk da0 and it returns *** Working on device /dev/da0 *** parameters extracted from in-core disklabel are: cylinders=8924 heads=255 sectors/track=63 (16065 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=8924 heads=255 sectors/track=63 (16065 blks/cyl) Media sector size is 512 Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 1 is: sysid 165,(FreeBSD/NetBSD/386BSD) start 63, size 143363997 (70001 Meg), flag 80 (active) beg: cyl 0/ head 1/ sector 1; end: cyl 1023/ head 254/ sector 63 The data for partition 2 is: The data for partition 3 is: The data for partition 4 is: So I tried: # disklabel da0s1 disklabel: ioctl DIOCGDINFO: Invalid argument # disklabel da0 disklabel: ioctl DIOCGDINFO: Invalid argument Is my disklabel missing, and that's why the raid-5 HDDs won't boot up and won't let me mount '/usr' and '/var'? Since the etc directory is accessible I can even access the old '/etc/fstab' and '/etc/disktab', would that help? All I need is to be able to mount /usr and /var and copy the data out to my IDE HD, whether the Raid-5 sys can ever boot up again is secondary. Here's '/etc/fstab' (well '/mnt/etc/fstab'): /dev/da0s1b noneswapsw 0 0 /dev/da0s1a / ufs rw 1 1 /dev/da0s1f /usrufs rw 2 2 /dev/da0s1e /varufs rw 2 2 /dev/da0s1g /wwwufs rw 2 2 /dev/acd0c /cdrom cd9660 ro,noauto 0 0 proc/proc procfs rw 0 0 Also, as a freaked out person desperatly trying to restore the data I tried to mount the slices. I tried: #mount /dev/da0s1f /mnt and it returns mount: /dev/da0s1f on /mnt: incorrect super block If anybody has any idea on how that can be done, please let me know; If any extra info is needed, please let me know and I'll provide them as much as possible. Thanks in advance. _ Get 10Mb extra storage for MSN Hotmail. Subscribe Now! http://join.msn.com/?pgmarket=en-hk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: No disklabel, but it still boots
Pete Yandell wrote: $ disklabel -r ad6 ~ disklabel: bad pack magic number (label is damaged, or pack is unlabeled) The disklabel is on ad6s1, not on ad6. The kernel does automatically generate "fictitious" labels for unlabeled disks, no matter if it's being used for BSD, or not. mkb. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
No disklabel, but it still boots
I have a disk which apparently has no disklabel, but still boots and mounts several partitions, and I'm puzzled as to how this can be. Where is it pulling the partition information from if not the disklabel? $ df Filesystem 1K-blocksUsed Avail Capacity Mounted on /dev/ad6s1a128990 48054 7061840%/ /dev/ad6s1f257998 18237342 0%/tmp /dev/ad6s1g 151150778 2131440 136927276 2%/usr /dev/ad6s1e2579987546229814 3%/var procfs 4 4 0 100%/proc $ disklabel ad6 # /dev/ad6: type: unknown disk: amnesiac label: fictitious flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 19457 sectors/unit: 312581808 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 3125818080unused0 0# (Cyl.0 - 19457*) $ disklabel -r ad6 ~ disklabel: bad pack magic number (label is damaged, or pack is unlabeled) If I boot off the fixit CD, I can't see or mount any FreeBSD partitions on the disk. This is on a HP Proliant DL360 Gen 4 with a single Maxtor 160GB SATA drive running FreeBSD 4.10. Thanks, Pete Yandell ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: install.cfg disklabel customization question
Good suggestion on using bsdlabel. Unfortunately I am required to use FreeBSD 4.6.2 which does not contain this utility and disklabel requires one to invoke an editor to define the new label. What I resorted to doing was having netboot create /usr100 and then later overwrite the /etc/fstab via an installation package that sets noauto for the label. Curtis On Wed, 05 Jan 2005 10:38:19 +, Matthew Seaman <[EMAIL PROTECTED]> wrote: > Curtis Almond wrote: > > > Anyone know how to make /usr100 not mounted at boot time? > > Edit /etc/fstab and add the 'noauto' flag to the appropriate line. > Something like this: > > /dev/ad0s2f /usr100 ufs rw,noauto 2 2 > > > Or even better > > How can I create the ad0s2-4 (ad0s2f after boot) label but have > > sysinstall not newfs it during netboot? > > You shouldn't need to recreate the disk or partition labels every time > you reboot, unless you are wiping and re-installing most of the disk > each time. > > If you're using sysinstall(8) to do an automatic install as part of your > netboot process, then as far as I can tell, there's no way using the > scripted interface to tell it to create a UFS partition but not newfs or > mount it -- although that's easy enough using sysinstall interactively. > > I'd be thinking more along the lines of ditching sysinstall(8) entirely > for that purpose and using fdisk(8), bsdlabel(8) and newfs(8) directly. > >Cheers, > >Matthew > > -- > Dr Matthew J Seaman MA, D.Phil. 8 Dane Court Manor > School Rd > PGP: http://www.infracaninophile.co.uk/pgpkey Tilmanstone > Tel: +44 1304 617253 Kent, CT14 0JL UK > > > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: install.cfg disklabel customization question
On Tue, 4 Jan 2005 17:23:05 -0600, Curtis Almond <[EMAIL PROTECTED]> wrote: > I would like to be able to do the following > > 1. Create a / partition of x size > 2. Create a swap partition of x size > 3. Create a /usr partition of x size > 4. Create a ufs partition of the rest of the disk but it is not mounted at > boot. > > What I have thus far is: > # label disk 1 > # IDE > ad0s2-1=ufs 3969000 / > ad0s2-2=swap 3969000 none > ad0s2-3=ufs 3969000 /usr > ad0s2-4=ufs 0 /usr100 > > Anyone know how to make /usr100 not mounted at boot time? Don't put it in /etc/fstab. Then you can use mount(8) to mount it when you need to. -- Joshua Lokken Open Source Advocate ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: install.cfg disklabel customization question
Curtis Almond wrote: Anyone know how to make /usr100 not mounted at boot time? Edit /etc/fstab and add the 'noauto' flag to the appropriate line. Something like this: /dev/ad0s2f /usr100 ufs rw,noauto 2 2 Or even better How can I create the ad0s2-4 (ad0s2f after boot) label but have sysinstall not newfs it during netboot? You shouldn't need to recreate the disk or partition labels every time you reboot, unless you are wiping and re-installing most of the disk each time. If you're using sysinstall(8) to do an automatic install as part of your netboot process, then as far as I can tell, there's no way using the scripted interface to tell it to create a UFS partition but not newfs or mount it -- although that's easy enough using sysinstall interactively. I'd be thinking more along the lines of ditching sysinstall(8) entirely for that purpose and using fdisk(8), bsdlabel(8) and newfs(8) directly. Cheers, Matthew -- Dr Matthew J Seaman MA, D.Phil. 8 Dane Court Manor School Rd PGP: http://www.infracaninophile.co.uk/pgpkey Tilmanstone Tel: +44 1304 617253 Kent, CT14 0JL UK signature.asc Description: OpenPGP digital signature
install.cfg disklabel customization question
I would like to be able to do the following 1. Create a / partition of x size 2. Create a swap partition of x size 3. Create a /usr partition of x size 4. Create a ufs partition of the rest of the disk but it is not mounted at boot. What I have thus far is: # label disk 1 # IDE ad0s2-1=ufs 3969000 / ad0s2-2=swap 3969000 none ad0s2-3=ufs 3969000 /usr ad0s2-4=ufs 0 /usr100 Anyone know how to make /usr100 not mounted at boot time? Or even better How can I create the ad0s2-4 (ad0s2f after boot) label but have sysinstall not newfs it during netboot? Any ideas would be great.. Curtis ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Identical hard drives, different disklabel sectors/cylinders
Thanks for the responses. Sorry for the delay in replying. orville weyrich said: > (1) make sure that the BIOS settings are the same for > both drives. I don't think this is an issue, as I haven't touched the BIOS for this IDE card, and it should be on full-auto mode. > (2) is there a jumper on the drive itself that affects > sector mapping? No. > (3) do you have all the /dev/ad4 and /dev/ad5 devices > (try MAKEDEV ad) I MAKEDEV'd both ad4 and ad5 at least once. > (4) have you tried the > /stand/sysinstall->configure->label method of editing > the labels? Does it report a discrepancy between BIOS > and memory disk geometry? Yes, I tried /stand/sysinstall several times, and actually I can't remember whether it gave me that error. However, I have gotten that error many times in the past and was under the impression that it was a fairly normal thing and should be ignored. > (5) have you tried using disklable -e to directly edit > the labels to copy the "good" label to the "bad" > drive? Yes, I tried editing both labels, and it wouldn't take for either. Christian Hiris said: > Your disks slice tables holding different values for C/H/S geometry. > You can wipe out the slice tables with the dd(1) command and > re-initialize them with fdisk and bsdlabel or sysinstall. There is an > example written on this in 'man 8 bsdlabel'. I dd'd both drives several times. I never used bsdlabel, as I was under the impression that disklabel is more recent. > IMHO there is only one thing that can go wrong with your configuration: > If you are booting your system from the mirror, I would test if both > of your drives are bootable. Well, these disks are just extra storage under /var, and ad0 and ad1 are the bootable disks in this system. So I guess I don't have anything to worry about. Again, thanks for your help :) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Identical hard drives, different disklabel sectors/cylinders
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sunday 19 December 2004 00:31, [EMAIL PROTECTED] wrote: > Hi all, > > I've got a problem question about my hard drives. I've installed > two brand-new Western Digital 250GB drives on the same channel of > a Maxtor-branded PCI/IDE controller (Promise chipset). I've tried > different channels, different cables, and using the motherboard's > IDE controller -- same thing every time. Here's from > /var/log/messages: > > Dec 18 15:59:23 /kernel: ad4: 238475MB > [484521/16/63] at ata2-master UDMA100 Dec 18 15:59:23 /kernel: ad5: > 238475MB [484521/16/63] at ata2-slave UDMA100 > > So you can see it's identifying them the same at startup. Now: > > su-2.05b# disklabel ad4 > # /dev/ad4c: > bytes/sector: 512 > sectors/track: 63 > tracks/cylinder: 255 > sectors/cylinder: 16065 > cylinders: 30400 > sectors/unit: 488392002 [...] > su-2.05b# disklabel ad5 > # /dev/ad5c: > bytes/sector: 512 > sectors/track: 63 > tracks/cylinder: 16 > sectors/cylinder: 1008 > cylinders: 484520 > sectors/unit: 488397105 [...] > So you can see, the "sectors/cylinder", "cylinders", and "sectors/unit" > are all different, and this results in slightly different final sizes. > I've set up vinum to create a mirror using the smaller of the two sizes, > and it seems to be working fine, but I'm still worried about the > implications of this. Could this mean that one of my hard drives is > failing? (Again, they're both brand new.) I tried "disklabel -R"'ing each > drive to look like the other one (booting in single-user mode), but it > wouldn't let me. So I'm smack out of ideas. I'd appreciate any info or > suggestions. Your disks slice tables holding different values for C/H/S geometry. You can wipe out the slice tables with the dd(1) command and re-initialize them with fdisk and bsdlabel or sysinstall. There is an example written on this in 'man 8 bsdlabel'. It seems that your BIOS identifies geometry of both disks listed in your dmesg output - C/H/S 484521/16/63. This values also shown in your bsdlabel output of ad5. If you want to re-fdisk and re-bsdlabel ad4 it's necessary that the disk isn't mounted and that it isn't locked by vinum (see also 'man 4 vinum', chapters RUNNING VINUM and AUTOMATIC STARTUP). IMHO there is only one thing that can go wrong with your configuration: If you are booting your system from the mirror, I would test if both of your drives are bootable. - -- Christian Hiris <[EMAIL PROTECTED]> | OpenPGP KeyID 0x3BCA53BE OpenPGP-Key at hkp://wwwkeys.eu.pgp.net and http://pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (FreeBSD) iD8DBQFBxNof09WjGjvKU74RAtgWAJ98M9KSoTPsHJU3Ba+JTHrNN2QdmwCfTmwy bFFRyDuNRfAwc3f1R/6PbeI= =j26F -END PGP SIGNATURE- ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Identical hard drives, different disklabel sectors/cylinders
Hi all, I've got a problem question about my hard drives. I've installed two brand-new Western Digital 250GB drives on the same channel of a Maxtor-branded PCI/IDE controller (Promise chipset). I've tried different channels, different cables, and using the motherboard's IDE controller -- same thing every time. Here's from /var/log/messages: Dec 18 15:59:23 /kernel: ad4: 238475MB [484521/16/63] at ata2-master UDMA100 Dec 18 15:59:23 /kernel: ad5: 238475MB [484521/16/63] at ata2-slave UDMA100 So you can see it's identifying them the same at startup. Now: su-2.05b# disklabel ad4 # /dev/ad4c: type: unknown disk: amnesiac label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 30400 sectors/unit: 488392002 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 4883920020unused0 0# (Cyl.0 - 30400*) e: 4883920020 vinum # (Cyl.0 - 30400*) su-2.05b# disklabel ad5 # /dev/ad5c: type: unknown disk: amnesiac label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 16 sectors/cylinder: 1008 cylinders: 484520 sectors/unit: 488397105 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 4883971050unused0 0# (Cyl.0 - 484520*) e: 4883971050 vinum # (Cyl.0 - 484520*) So you can see, the "sectors/cylinder", "cylinders", and "sectors/unit" are all different, and this results in slightly different final sizes. I've set up vinum to create a mirror using the smaller of the two sizes, and it seems to be working fine, but I'm still worried about the implications of this. Could this mean that one of my hard drives is failing? (Again, they're both brand new.) I tried "disklabel -R"'ing each drive to look like the other one (booting in single-user mode), but it wouldn't let me. So I'm smack out of ideas. I'd appreciate any info or suggestions. Thanks, -Brock Witherspoon ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: switching ide disk, change disklabel?
On Fri, 3 Dec 2004 04:07 am, Alex Teslik wrote: > > If you mean to move the disk containing ads1s1e from primary slave to the > > primary master then the partition will automatically become ad0s1e. > > Great. That answers my question exactly. > > > But how are the other partitions in your system organised? Are you > > currently using ad0s1a, ad0s1b etc.? > > FilesystemSize Used Avail Capacity Mounted on > /dev/da0s1a97M62M27M70%/ > /dev/da0s1e 7.7G 6.2G 900M88%/usr > /dev/ad0s1e72G66G 772M99%/home > procfs4.0K 4.0K 0B 100%/proc > > So I'm booting off of a SCSI disk da0. > OK > I want to replace ad0s1e with the new 300GB monster. > > > These will move with the disk. If you are physically > > swapping the disk connections these will become ad1s1a, ad1s1b etc. which > > will be a problem. You'll need to fix /etc/fstab and arrange for a > > booting MBR on the new disk and arrange for the boot sequence to find > > your / partition. > > But not in my case since this disk is just storage I think, right. > Actually, when I run the disklabel do I need to use -B at all. I don't need > a bootstrap since its not a boot disk, right? > Right Malcolm ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: switching ide disk, change disklabel?
> If you mean to move the disk containing ads1s1e from primary slave to the > primary master then the partition will automatically become ad0s1e. Great. That answers my question exactly. > > But how are the other partitions in your system organised? Are you currently > using ad0s1a, ad0s1b etc.? FilesystemSize Used Avail Capacity Mounted on /dev/da0s1a97M62M27M70%/ /dev/da0s1e 7.7G 6.2G 900M88%/usr /dev/ad0s1e72G66G 772M99%/home procfs4.0K 4.0K 0B 100%/proc So I'm booting off of a SCSI disk da0. I want to replace ad0s1e with the new 300GB monster. > These will move with the disk. If you are physically > swapping the disk connections these will become ad1s1a, ad1s1b etc. which will > be a problem. You'll need to fix /etc/fstab and arrange for a booting MBR on > the new disk and arrange for the boot sequence to find your / partition. But not in my case since this disk is just storage I think, right. Actually, when I run the disklabel do I need to use -B at all. I don't need a bootstrap since its not a boot disk, right? Thanks, Alex ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: switching ide disk, change disklabel?
On Thu, 2 Dec 2004 06:38 pm, Alex Teslik wrote: > Hello, > >I am adding a new disk to the system to make my /home partition bigger. > Currently I have /home on ad0s1e. I will be adding ad1s1e. After I copy all > the data from ad0 to ad1 I want to remove ad0 and make ad1 the master on > that controller. Will I need to change the disklabel of ad1 to ad0 at that > point, or will FreeBSD automagically know what to do? If you mean to move the disk containing ads1s1e from primary slave to the primary master then the partition will automatically become ad0s1e. But how are the other partitions in your system organised? Are you currently using ad0s1a, ad0s1b etc.? These will move with the disk. If you are physically swapping the disk connections these will become ad1s1a, ad1s1b etc. which will be a problem. You'll need to fix /etc/fstab and arrange for a booting MBR on the new disk and arrange for the boot sequence to find your / partition. Malcolm ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
switching ide disk, change disklabel?
Hello, I am adding a new disk to the system to make my /home partition bigger. Currently I have /home on ad0s1e. I will be adding ad1s1e. After I copy all the data from ad0 to ad1 I want to remove ad0 and make ad1 the master on that controller. Will I need to change the disklabel of ad1 to ad0 at that point, or will FreeBSD automagically know what to do? Note that this is all on the second controller, so no booting happening here. Just /home data. uname -a FreeBSD xx..com 4.9-RELEASE-p2 FreeBSD 4.9-RELEASE-p2 #1: Tue Jun 8 19:15:51 PDT 2004 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/DUALP3-RELENG_4_9 i386 Thanks, Alex ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: vinum disklabel FBSD 5.2.1....
On 07 nov 2004, at 00:19, Greg 'groggy' Lehey wrote: On Sunday, 31 October 2004 at 14:03:18 +0100, FreeBSD questions mailing list wrote: On 31 okt 2004, at 07:41, matt virus wrote: matt virus wrote: Hi all! I have (8) maxtor 160gb drives I plan on constructing a vinum raid5 array with. the devices are: ad4ad11 All drives have been fdisk'd and such, ad4s1d.ad11s1d The first step of setting up vinum is changing the disklabel disklabel -e /dev/ad4 The disk label says it has 8 partitions, but only the A and C partitions are shown... ******MY DISKLABEL # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 320173040 16unused0 0 c: 3201730560unused0 0 # "raw" part, don't edit ** c: is not a valid disk label. You need to create one first. See the example below first: there's an "e" label. You can do this in sysinstall: Configure / Label / ad4 and then C to create one. Once that's done it'll show up in disklabel as you write below. Then in disklabel you can change the 4.2BSD to vinum. You should also not use 'c' for Vinum. Greg -- Bit of confusion from my side: I meant C as the key that should be pressed to create a new slice not as a name for a disklabel Arno ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: vinum disklabel FBSD 5.2.1....
On Sunday, 31 October 2004 at 14:03:18 +0100, FreeBSD questions mailing list wrote: > > On 31 okt 2004, at 07:41, matt virus wrote: > >> matt virus wrote: >>> Hi all! >>> I have (8) maxtor 160gb drives I plan on constructing a vinum raid5 >>> array with. >>> the devices are: >>> ad4ad11 >>> All drives have been fdisk'd and such, >>> ad4s1d.ad11s1d >>> The first step of setting up vinum is changing the disklabel >>> disklabel -e /dev/ad4 >>> The disk label says it has 8 partitions, but only the A and C >>> partitions are shown... >>> **MY DISKLABEL >>> # /dev/ad4: >>> 8 partitions: >>> #size offsetfstype [fsize bsize bps/cpg] >>> a: 320173040 16unused0 0 >>> c: 3201730560unused0 0 # "raw" part, don't edit >>> ** > > c: is not a valid disk label. You need to create one first. See the > example below first: there's an "e" label. You can do this in > sysinstall: Configure / Label / ad4 and then C to create one. Once > that's done it'll show up in disklabel as you write below. Then in > disklabel you can change the 4.2BSD to vinum. You should also not use 'c' for Vinum. Greg -- When replying to this message, please copy the original recipients. If you don't, I may ignore the reply or reply to the original recipients. For more information, see http://www.lemis.com/questions.html See complete headers for address and phone numbers. pgpP1ctMiCuRw.pgp Description: PGP signature
Re: vinum disklabel FBSD 5.2.1....
On 31 okt 2004, at 07:41, matt virus wrote: nobody ??? OK I'll give it a try. I have a vinum RAID 1 running though, but the way to get it tunning isn't very different. matt virus wrote: Hi all! I have (8) maxtor 160gb drives I plan on constructing a vinum raid5 array with. the devices are: ad4ad11 All drives have been fdisk'd and such, ad4s1d.ad11s1d The first step of setting up vinum is changing the disklabel disklabel -e /dev/ad4 The disk label says it has 8 partitions, but only the A and C partitions are shown... ******MY DISKLABEL # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 320173040 16unused0 0 c: 3201730560unused0 0 # "raw" part, don't edit ** c: is not a valid disk label. You need to create one first. See the example below first: there's an "e" label. You can do this in sysinstall: Configure / Label / ad4 and then C to create one. Once that's done it'll show up in disklabel as you write below. Then in disklabel you can change the 4.2BSD to vinum. Of course you can also add the whole label-line in disklabel itself but I find sysinstall easier. Arno Now, i know i have to change *something* to "vinum" but i'm unsure which one, or if I need to actually add a line or ??? This is my first time playing with vinum, i've read a handful of howtos and all the documentation I find shows the disklabel looking like this: *HOWTO's Disklabel # disklabel da0 [snip] #size offsetfstype [fsize bsize bps/cpg] a: 1024000 10240004.2BSD 2048 1638490 b: 10240000 swap c: 179124120unused0 0 e: 15864412 2048000 vinum (source: http://org.netbase.org/vinum-mirrored.html) Any direction is appreciated :-) -matt ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: vinum disklabel FBSD 5.2.1....
nobody ??? matt virus wrote: Hi all! I have (8) maxtor 160gb drives I plan on constructing a vinum raid5 array with. the devices are: ad4ad11 All drives have been fdisk'd and such, ad4s1d.ad11s1d The first step of setting up vinum is changing the disklabel disklabel -e /dev/ad4 The disk label says it has 8 partitions, but only the A and C partitions are shown... **MY DISKLABEL # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 320173040 16unused0 0 c: 3201730560unused0 0 # "raw" part, don't edit ** Now, i know i have to change *something* to "vinum" but i'm unsure which one, or if I need to actually add a line or ??? This is my first time playing with vinum, i've read a handful of howtos and all the documentation I find shows the disklabel looking like this: *****HOWTO's Disklabel # disklabel da0 [snip] #size offsetfstype [fsize bsize bps/cpg] a: 1024000 10240004.2BSD 2048 1638490 b: 10240000 swap c: 179124120unused0 0 e: 15864412 2048000 vinum (source: http://org.netbase.org/vinum-mirrored.html) Any direction is appreciated :-) -matt ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" -- Matt Virus ("veer-iss") www.mattvirus.net ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
vinum disklabel FBSD 5.2.1....
Hi all! I have (8) maxtor 160gb drives I plan on constructing a vinum raid5 array with. the devices are: ad4ad11 All drives have been fdisk'd and such, ad4s1d.ad11s1d The first step of setting up vinum is changing the disklabel disklabel -e /dev/ad4 The disk label says it has 8 partitions, but only the A and C partitions are shown... **MY DISKLABEL # /dev/ad4: 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 320173040 16unused0 0 c: 3201730560unused0 0 # "raw" part, don't edit ** Now, i know i have to change *something* to "vinum" but i'm unsure which one, or if I need to actually add a line or ??? This is my first time playing with vinum, i've read a handful of howtos and all the documentation I find shows the disklabel looking like this: *****HOWTO's Disklabel # disklabel da0 [snip] #size offsetfstype [fsize bsize bps/cpg] a: 1024000 10240004.2BSD 2048 1638490 b: 10240000 swap c: 179124120unused0 0 e: 15864412 2048000 vinum (source: http://org.netbase.org/vinum-mirrored.html) Any direction is appreciated :-) -matt ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: corrupt disklabel, how to restore?
Ok, thanks I managed to rescue my disk from /usr/sbin/sysinstall Freebsd Disklabel Editor in combination with fdisk. It automatically fsck'd my filesystem in write of disklabel... Thanks :) -d Subhro wrote: Y means a new file system would be created Regards S. On Wed, 29 Sep 2004 11:45:26 +0200, didi <[EMAIL PROTECTED]> wrote: Does : UFS2+S N in "freebsd disklabel editor" mean that no newfs will be created i.e that only the disklabel will be written and the acyual filesystem be left alone? as opposed to UFS+S Y ? d-tail (I'm a bit worried :| ) Subhro wrote: There is a key for toggling new filesystem. Probably Y. sorry don't rember it. Regards S. On Wed, 29 Sep 2004 09:58:07 +0200, didi <[EMAIL PROTECTED]> wrote: It's partitioned and mounted from sysinstall in 5.2.1 And the machine rins 5.2.1 again :( So, I guess I could: use fdisk and/or disklabel from within sysinstall again. But how do I prevent sysinstall from making 'newfs' ? Thanks & Regards D-tail Subhro wrote: If you have partitioned your disk as UFS2, then there is no way in which you can access them running 4.10. Regards S, On Wed, 29 Sep 2004 08:32:18 +0200, didi <[EMAIL PROTECTED]> wrote: I did what you shouldn't do. I have two disks one 80gb and one 100gb I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. Problem is the disklabel on my second drive somehow got corrupt, on this drive i backed up all home/* folders etc and so on, but now I can't get to my backup. I really need a way to restore the disklabel. I have tried scan_ffs with no result is there any other way, The disk was mounted as ad1s1d one large slice. Any information could be usefull ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: corrupt disklabel, how to restore?
Y means a new file system would be created Regards S. On Wed, 29 Sep 2004 11:45:26 +0200, didi <[EMAIL PROTECTED]> wrote: > > Does : > > UFS2+S N > > in "freebsd disklabel editor" > > mean that no newfs will be created i.e that only the disklabel will be > written and the acyual filesystem be left alone? > > as opposed to > > UFS+S Y > > ? > > d-tail > > (I'm a bit worried :| ) > > > > > Subhro wrote: > > There is a key for toggling new filesystem. Probably Y. sorry > > don't rember it. > > > > Regards > > S. > > > > > > On Wed, 29 Sep 2004 09:58:07 +0200, didi <[EMAIL PROTECTED]> wrote: > > > >>It's partitioned and mounted from sysinstall in 5.2.1 > >> > >>And the machine rins 5.2.1 again :( > >> > >>So, I guess I could: > >> > >>use fdisk and/or disklabel from within sysinstall again. But how do I > >>prevent sysinstall from making 'newfs' ? > >> > >>Thanks & Regards D-tail > >> > >> > >> > >> > >>Subhro wrote: > >> > >>>If you have partitioned your disk as UFS2, then there is no way in > >>>which you can access them running 4.10. > >>> > >>>Regards > >>>S, > >>> > >>> > >>>On Wed, 29 Sep 2004 08:32:18 +0200, didi <[EMAIL PROTECTED]> wrote: > >>> > >>> > >>>>I did what you shouldn't do. I have two disks one 80gb and one 100gb > >>>>I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. > >>>>Problem is the disklabel on my second drive somehow got corrupt, on this > >>>>drive i backed up all home/* folders etc and so on, but now I can't get > >>>>to my backup. > >>>>I really need a way to restore the disklabel. > >>>> > >>>>I have tried scan_ffs with no result is there any other way, > >>>> > >>>>The disk was mounted as ad1s1d one large slice. > >>>> > >>>>Any information could be usefull > >>>>___ > >>>>[EMAIL PROTECTED] mailing list > >>>>http://lists.freebsd.org/mailman/listinfo/freebsd-questions > >>>>To unsubscribe, send any mail to "[EMAIL PROTECTED]" > >>>> > >>> > >>> > >>> > >>> > > > > > > > -- Subhro Sankha Kar School of Information Technology Block AQ-13/1 Sector V ZIP 700091 India ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: corrupt disklabel, how to restore?
Does : UFS2+S N in "freebsd disklabel editor" mean that no newfs will be created i.e that only the disklabel will be written and the acyual filesystem be left alone? as opposed to UFS+S Y ? d-tail (I'm a bit worried :| ) Subhro wrote: There is a key for toggling new filesystem. Probably Y. sorry don't rember it. Regards S. On Wed, 29 Sep 2004 09:58:07 +0200, didi <[EMAIL PROTECTED]> wrote: It's partitioned and mounted from sysinstall in 5.2.1 And the machine rins 5.2.1 again :( So, I guess I could: use fdisk and/or disklabel from within sysinstall again. But how do I prevent sysinstall from making 'newfs' ? Thanks & Regards D-tail Subhro wrote: If you have partitioned your disk as UFS2, then there is no way in which you can access them running 4.10. Regards S, On Wed, 29 Sep 2004 08:32:18 +0200, didi <[EMAIL PROTECTED]> wrote: I did what you shouldn't do. I have two disks one 80gb and one 100gb I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. Problem is the disklabel on my second drive somehow got corrupt, on this drive i backed up all home/* folders etc and so on, but now I can't get to my backup. I really need a way to restore the disklabel. I have tried scan_ffs with no result is there any other way, The disk was mounted as ad1s1d one large slice. Any information could be usefull ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: corrupt disklabel, how to restore?
There is a key for toggling new filesystem. Probably Y. sorry don't rember it. Regards S. On Wed, 29 Sep 2004 09:58:07 +0200, didi <[EMAIL PROTECTED]> wrote: > It's partitioned and mounted from sysinstall in 5.2.1 > > And the machine rins 5.2.1 again :( > > So, I guess I could: > > use fdisk and/or disklabel from within sysinstall again. But how do I > prevent sysinstall from making 'newfs' ? > > Thanks & Regards D-tail > > > > > Subhro wrote: > > If you have partitioned your disk as UFS2, then there is no way in > > which you can access them running 4.10. > > > > Regards > > S, > > > > > > On Wed, 29 Sep 2004 08:32:18 +0200, didi <[EMAIL PROTECTED]> wrote: > > > >>I did what you shouldn't do. I have two disks one 80gb and one 100gb > >>I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. > >>Problem is the disklabel on my second drive somehow got corrupt, on this > >>drive i backed up all home/* folders etc and so on, but now I can't get > >>to my backup. > >>I really need a way to restore the disklabel. > >> > >>I have tried scan_ffs with no result is there any other way, > >> > >>The disk was mounted as ad1s1d one large slice. > >> > >>Any information could be usefull > >>___ > >>[EMAIL PROTECTED] mailing list > >>http://lists.freebsd.org/mailman/listinfo/freebsd-questions > >>To unsubscribe, send any mail to "[EMAIL PROTECTED]" > >> > > > > > > > > > -- Subhro Sankha Kar School of Information Technology Block AQ-13/1 Sector V ZIP 700091 India ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: corrupt disklabel, how to restore?
It's partitioned and mounted from sysinstall in 5.2.1 And the machine rins 5.2.1 again :( So, I guess I could: use fdisk and/or disklabel from within sysinstall again. But how do I prevent sysinstall from making 'newfs' ? Thanks & Regards D-tail Subhro wrote: If you have partitioned your disk as UFS2, then there is no way in which you can access them running 4.10. Regards S, On Wed, 29 Sep 2004 08:32:18 +0200, didi <[EMAIL PROTECTED]> wrote: I did what you shouldn't do. I have two disks one 80gb and one 100gb I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. Problem is the disklabel on my second drive somehow got corrupt, on this drive i backed up all home/* folders etc and so on, but now I can't get to my backup. I really need a way to restore the disklabel. I have tried scan_ffs with no result is there any other way, The disk was mounted as ad1s1d one large slice. Any information could be usefull ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: corrupt disklabel, how to restore?
If you have partitioned your disk as UFS2, then there is no way in which you can access them running 4.10. Regards S, On Wed, 29 Sep 2004 08:32:18 +0200, didi <[EMAIL PROTECTED]> wrote: > I did what you shouldn't do. I have two disks one 80gb and one 100gb > I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. > Problem is the disklabel on my second drive somehow got corrupt, on this > drive i backed up all home/* folders etc and so on, but now I can't get > to my backup. > I really need a way to restore the disklabel. > > I have tried scan_ffs with no result is there any other way, > > The disk was mounted as ad1s1d one large slice. > > Any information could be usefull > ___ > [EMAIL PROTECTED] mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" > -- Subhro Sankha Kar School of Information Technology Block AQ-13/1 Sector V ZIP 700091 India ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
corrupt disklabel, how to restore?
I did what you shouldn't do. I have two disks one 80gb and one 100gb I was not happy with FreeBSD 5.2.1 so I downgraded to 4.10. Problem is the disklabel on my second drive somehow got corrupt, on this drive i backed up all home/* folders etc and so on, but now I can't get to my backup. I really need a way to restore the disklabel. I have tried scan_ffs with no result is there any other way, The disk was mounted as ad1s1d one large slice. Any information could be usefull ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel
Hi all, I have 6 GB left on an already production disk drive, that I want to label and use. In sysinstall, using disklabel, I get the following: Disk: ar0 Partition name: ar0s1 Free: 12218275 blocks (5965MB) Part Mount Size Newfs Part Mount Size Newfs - - - - ar0s1a250MB * ar0s1bswap 1024MB SWAP ar0s1e 4MB * ar0s1f 3MB * ar0s1g 4MB * If I create another partition and then write with the 'w' command, will this destroy any data? Or would I be better off using the disklabel command, and entering in the following: h: * 4.2BSD and then doing a newfs? What is the safest or most recommended way? Tks, Steve ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
install.cfg disklabel failure when PXE booting
I am attempting to set up network installs of FreeBSD 5.2.1 for a project I am working on. I can boot kern.flp and mfsroot.flp over the network with no problems and do a manual install. I can configure install.cfg to automatically set the distribution, network card, install method and so on. What I can not get working right is automatic disk partitioning. No matter what I have set up, the installer just will not set up partitions for me. install.cfg looks like: # Now set the parameters for the partition editor on ad0 disk=ad0 partition=all bootManager=standard diskPartitionEditor #diskPartitionWrite # All sizes are expressed in 512 byte blocks! ad0s1-1=ufs 199 / ad0s1-2=swap 6485760 none ad0s1-3=ufs 2097152 /var ad0s1-4=ufs 0 /usr # Let's do it! diskLabelEditor #diskLabelCommit Can anyone tell me where I am screwing up? I have tried several variations on the above config file all without success. There is also the somewhat related problem of sysinstall refusing to accept the BIOS definition of the drive geometry but I believe I was able to get beyond that problem. Thanks in advance, -Don ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel question and enlarging a diskslice
On Sat, 17 Jan 2004 15:22:40 +0200 Ion-Mihai Tetcu <[EMAIL PROTECTED]> wrote: > See growfs(8). All ready figured out a way of moving stuff around to fix the prob =] btw growfs is not useful here becuase it is a diskslice that was the prob, not a fs. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel question and enlarging a diskslice
On Fri, 16 Jan 2004 17:20:09 -0600 kitsune <[EMAIL PROTECTED]> wrote: > On Fri, 16 Jan 2004 17:05:35 -0600 > kitsune <[EMAIL PROTECTED]> wrote: > > > Is it possible to use disklabel to enlarge a diskslice? > > > > I have been doing a fresh install of fbsd and I > > forgot I have a 40GB instead of a 20GB drive. Can this be fixed > > using disklabel or do I have to copy everything to another drive, > > reslice it, and then copy the stuff back over? > > Whoops, this question should have been over fdisk, not See growfs(8). -- IOnut Unregistered ;) FreeBSD user ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel question and enlarging a diskslice
On Fri, 16 Jan 2004 17:05:35 -0600 kitsune <[EMAIL PROTECTED]> wrote: > Is it possible to use disklabel to enlarge a diskslice? > > I have been doing a fresh install of fbsd and I > forgot I have a 40GB instead of a 20GB drive. Can this be fixed > using disklabel or do I have to copy everything to another drive, > reslice it, and then copy the stuff back over? Whoops, this question should have been over fdisk, not disklabel. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel question and enlarging a diskslice
Is it possible to use disklabel to enlarge a diskslice? I have been doing a fresh install of fbsd and I forgot I have a 40GB instead of a 20GB drive. Can this be fixed using disklabel or do I have to copy everything to another drive, reslice it, and then copy the stuff back over? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: disklabel question
On Sat, Dec 27, 2003 at 09:02:37AM -0500, Jason wrote: > hey folks, I just migrated my freebsd server from a couple of > older/smaller scsi disks to a new ide controller with a bigger ATA133 > disk. I have all the partitions moved over to the new drive.. > the /etc/fstab on the new drive is updated. The bootmanager is installed > on the new drive.. but when I boot, I get > > Invalid partition > No /boot/loader > > >>FreeBSD/i386 BOOT > Default: 0:ad(0,a)/kernel > boot: > > At the "boot:" prompt if I enter "0:ad(4,e)/kernel", it boots fine. > > here is the old vs the new > FilesystemSize Used Avail Capacity Mounted on > /dev/da0s1a 126M53M63M45%/ > /dev/da0s1f 252M21M 211M 9%/tmp > /dev/da0s1e 252M 124M 108M53%/var > /dev/da1s1e 3.9G 2.7G 946M74%/usr > procfs4.0K 4.0K 0B 100%/proc > > new: > FilesystemSize Used Avail Capacity Mounted on > /dev/ad4s1e 984M 534M 372M59%/ > /dev/ad4s1f36G 2.2G31G 6%/usr > procfs4.0K 4.0K 0B 100%/proc > > > so Im thinking I need to do something with disklabel but I want to make > sure that I dont destroy my "so far working" efforts. > > regards, > Jason I believe a look through /boot/defaults/loader.conf should set you on the path to booting goodness. :) Josh Paetzel ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
disklabel question
hey folks, I just migrated my freebsd server from a couple of older/smaller scsi disks to a new ide controller with a bigger ATA133 disk. I have all the partitions moved over to the new drive.. the /etc/fstab on the new drive is updated. The bootmanager is installed on the new drive.. but when I boot, I get Invalid partition No /boot/loader >>FreeBSD/i386 BOOT Default: 0:ad(0,a)/kernel boot: At the "boot:" prompt if I enter "0:ad(4,e)/kernel", it boots fine. here is the old vs the new FilesystemSize Used Avail Capacity Mounted on /dev/da0s1a 126M53M63M45%/ /dev/da0s1f 252M21M 211M 9%/tmp /dev/da0s1e 252M 124M 108M53%/var /dev/da1s1e 3.9G 2.7G 946M74%/usr procfs4.0K 4.0K 0B 100%/proc new: FilesystemSize Used Avail Capacity Mounted on /dev/ad4s1e 984M 534M 372M59%/ /dev/ad4s1f36G 2.2G31G 6%/usr procfs4.0K 4.0K 0B 100%/proc so Im thinking I need to do something with disklabel but I want to make sure that I dont destroy my "so far working" efforts. regards, Jason ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Disklabel again
Hi, I am reposting this in hopes poeple who have disklabel and RAID perc 3DCL / SCSI experience see it tonight. Hi, I had a crash tonight. Server rebooted and everything seemed to restart OK. Interesting disklabel output though. Should I be worried about all the "*"s? : If this helps, the disk below is a 10,000 spin RAID 5 Dell Perc DC/L (3 Fujitu disks). enterprise# disklabel /dev/amrd0 # /dev/amrd0: type: SCSI disk: amnesiac label: fictitious flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 4405 sectors/unit: 70770688 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 707706880unused0 0 # (Cyl.0 - 4405*) enterprise# enterprise# disklabel /dev/amrd0s1 # /dev/amrd0s1: type: ESDI disk: amrd0s1 label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 4404 sectors/unit: 70766262 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 102400004.2BSD0 0 0 # (Cyl.0 - 63*) b: 2048000 1024000 swap# (Cyl. 63*- 191*) c: 707662620unused0 0 # (Cyl.0 - 4404*) d: 8142848 461291524.2BSD 1024 819216 # (Cyl. 2871*- 3378*) e: 1024 30720004.2BSD0 0 0 # (Cyl. 191*- 828*) f: 3072 133120004.2BSD0 0 0 # (Cyl. 828*- 2740*) g: 2097152 440320004.2BSD0 0 0 # (Cyl. 2740*- 2871*) h: 16494262 542720004.2BSD0 0 0 # (Cyl. 3378*- 4404*) enterprise# ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Disklabel
Hi, I had a crash tonight. Server rebooted and everything seemed to restart OK. Interesting disklabel output though. Should I be worried about all the "*"s? : If this helps, the disk below is a 10,000 spin RAID 5 Dell Perc DC/L (3 Fujitu disks). enterprise# disklabel /dev/amrd0 # /dev/amrd0: type: SCSI disk: amnesiac label: fictitious flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 4405 sectors/unit: 70770688 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 707706880unused0 0 # (Cyl.0 - 4405*) enterprise# enterprise# disklabel /dev/amrd0s1 # /dev/amrd0s1: type: ESDI disk: amrd0s1 label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 4404 sectors/unit: 70766262 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] a: 102400004.2BSD0 0 0 # (Cyl.0 - 63*) b: 2048000 1024000 swap# (Cyl. 63*- 191*) c: 707662620unused0 0 # (Cyl.0 - 4404*) d: 8142848 461291524.2BSD 1024 819216 # (Cyl. 2871*- 3378*) e: 1024 30720004.2BSD0 0 0 # (Cyl. 191*- 828*) f: 3072 133120004.2BSD0 0 0 # (Cyl. 828*- 2740*) g: 2097152 440320004.2BSD0 0 0 # (Cyl. 2740*- 2871*) h: 16494262 542720004.2BSD0 0 0 # (Cyl. 3378*- 4404*) enterprise# ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disklabel problem IBM SCSI3 disks, vinum too
Date sent: Mon, 24 Nov 2003 09:16:49 +1030 From: "Greg 'groggy' Lehey" <[EMAIL PROTECTED]> To: "Bob Collins" <[EMAIL PROTECTED]> Organization: The FreeBSD Project Copies to: FreeBSD <[EMAIL PROTECTED]> Subject:Re: Disklabel problem IBM SCSI3 disks, vinum too [Format recovered--see http://www.lemis.com/email/email-format.html] Single-line paragraphs. On Sunday, 23 November 2003 at 13:03:47 -0500, Bob Collins wrote: > At 07:30 PM 11/21/2003, you wrote: >> On Friday, 21 November 2003 at 9:25:58 -0500, Bob Collins wrote: >>> vinum -> list >>> 4 drives: >>> D d State: up Device /dev/da1s1h Avail: 0/8747 MB >>> (0%) >>> D c State: up Device /dev/da2s1h Avail: 0/8747 MB >>> (0%) >>> D b State: up Device /dev/da3s1h Avail: 0/8747 MB >>> (0%) >>> D a State: up Device /dev/da4s1h Avail: 0/8747 MB >>> (0%) >>> >>> 1 volumes: >>> V raid State: down Plexes: 1 Size: 25 GB >>> >>> 1 plexes: >>> P raid.p0R5 State: init Subdisks: 4 Size: 25 GB >>> >>> 4 subdisks: >>> S raid.p0.s0State: emptyPO:0 B Size: 8747 MB >>> S raid.p0.s1State: emptyPO: 512 kB Size: 8747 MB >>> S raid.p0.s2State: emptyPO: 1024 kB Size: 8747 MB >>> S raid.p0.s3State: emptyPO: 1536 kB Size: 8747 MB >> >> This doesn't agree with what you say above. It also looks fine to >> me. > > My apologies, Greg. Quite right. I was messing with the system right > before the response about sa4 device nodes in /dev. Once I made the > da4 devices, then the disklabel worked. My mistake jumping the gun. > > One follow-up question if I may. > > I assume that the init process for a RAID5 takes quite some time, > no? This has been in the init stage for 3 days. The vinum daemon is > running as I can see it listed in ps -ax. This is the init state, which means it needs initializing. To initialize, issue the 'init' command. The state will change from 'init' to 'initializing', and the list command shows the progress of the initialization. This writes zeroes to the drives in parallel; expect it to take an hour or two on drives of this size. Greg -- Thanks Greg, that did the trick. The initialization only took 19 minutes. The newfs took about 5 minutes, and it is now running perfectly. Thank you again. Bob -- Bob Collins Anything-inc.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"