Re: Windows support in 2.4.2p2 vs 2.4.1p2?
Jon, >I'm very nieve about Windows. 25 years unix, 0.5 year win2K. Forget it, while one can aquire experience about Unix, one can never about Microsoft. Well yes, one can acquire ONE experience about Microsoft, that is "there is nothing to be learned about that GUI interruption handler" After 25 years of Unix, I beleive the experience is unbearably painfull for you :) Why not installin FreeBSD on that box :) OK more serious: >From your Amanda/samba box, you should try something like smbclient //machine/share -d 0 -U backup -E -W domain -c "archive 1;recurse;du" or smbclient //machine/share -d 0 -U backup%password -E -W domain -c "archive 1;recurse;du" As far as I understood, a W2000 machine is always in a domain, even if it is all by itself in that domain. On the W2K mnachine control pannel/system/network identification should tell you the thing. Once one of the above command is working, you have it configured, then put in /etc/amandapass //machine/share backup%password domain Be patient, it does work after a while. Don't ask me the rational for all that, there are none, we are talking about Micrisoft. Olivier
Re: Looking for a little help
On Thu, Jun 21, 2001 at 05:36:19PM -0600, Kevin C Jones ([EMAIL PROTECTED]) wrote: > I have installed amanda onto a Redhat 6.2 box. > Everything went into okay, I edited the amanda.conf and made a dumptype and > disklist. > > When I run the amcheck Daily comment... I get the following error: [snip] > Amanda Backup Client Hosts Check > > ERROR: : [access as amanda not allowed from is here>: amandahostsauth failed > Client check: 1 host checked in 0.012 seconds, 1 problem found You forgot to put the ".amandahosts" file into the amanda home directory which should have an entry FOR each client AND username: file .amandahosts (one line for each) examplehostname.in.domain.com amanda Note that the user and machine name are separated by a TAB! jobst -- We're from the government, we're here to help you... |__, Jobst Schmalenbach, [EMAIL PROTECTED], Technical Director| | _ _.--'-n_/ Barrett Consulting Group P/L & The Meditation Room P/L | |-(_)--(_)= +61 3 9532 7677, POBox 277, Caulfield South, 3162, Australia|
Re: Windows support in 2.4.2p2 vs 2.4.1p2?
>I want to use Amanda to backup my Windows laptop disk. On the Windows >side, I'd rather share the disk with user-based authentication than a >simple share-level password. However I don't control the domain >controller, so I can't create a user 'backup' and would rather use my >own username. However I'm running 2.4.1p2 and it seems to force the >username to 'backup'. What are you running on your laptop? I mean WHAT windows? It makes big differences. And what Samba have you installed? This makes differences too I understood. >From experience, on Win2000, amandapass must be something like //machine/share user%password domain It must have a domain name Olivier
selfcheck request timed out - redux
When I run amcheck I get 'selfcheck request timed out' while trying to connect to localhost on a new installation on a NetBSD machine I'm checking out. NetBSD 1.5 on a Mac Quadra 660av with an Exabyte 8500 drive; in case it matters. I remember getting this error on other installations. It seems to get me every time I do a new install or even an update on an old machine. >From time to time it is the missed typo in inetd.conf or permissions in /etc/osts.allow. Other times it is something truly mysterious to me, that I can ignore because I usually manage to get it working. I have found that with a machine with two ethernet interfaces, each with its own IP and name, one will work and the other will fail. I have found cases where I need to put both the user and the operator in inetd.conf. Does anyone have a complete list of what causes this error message? I keep thinking about what it could be and then I try something. If it fixes it I stop (and forget exactly what I did to get it working). I have seen notes about using lsof to debug, but that is tricky when the client and server are on the same machine. -- Josh Kuperman [EMAIL PROTECTED]
Looking for a little help
I have installed amanda onto a Redhat 6.2 box. Everything went into okay, I edited the amanda.conf and made a dumptype and disklist. When I run the amcheck Daily comment... I get the following error: /usr/local/sbin/amcheck Daily Amanda Tape Server Host Check - Holding disk /home/users/amanda: 2372476 KB disk space available, that's plenty NOTE: skipping tape-writable test Tape Daily08 label ok NOTE: info dir /www/logs/amanda/Daily/curinfo/: does not exist Server check took 4.924 seconds Amanda Backup Client Hosts Check ERROR: : [access as amanda not allowed from : amandahostsauth failed Client check: 1 host checked in 0.012 seconds, 1 problem found (brought to you by Amanda 2.4.2p2) Any ideas on where I can look to try and correct this? KJ
Re: Windows support in 2.4.2p2 vs 2.4.1p2?
On Thu, 21 Jun 2001 at 5:41pm, David E. Bernholdt wrote > I want to use Amanda to backup my Windows laptop disk. On the Windows > side, I'd rather share the disk with user-based authentication than a > simple share-level password. However I don't control the domain > controller, so I can't create a user 'backup' and would rather use my > own username. However I'm running 2.4.1p2 and it seems to force the > username to 'backup'. > It should use whatever username is in /etc/amandapass on the samba host. My /etc/amandapass looks like: //pathto/shareUSER%PASS -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Windows support in 2.4.2p2 vs 2.4.1p2?
I want to use Amanda to backup my Windows laptop disk. On the Windows side, I'd rather share the disk with user-based authentication than a simple share-level password. However I don't control the domain controller, so I can't create a user 'backup' and would rather use my own username. However I'm running 2.4.1p2 and it seems to force the username to 'backup'. I've been looking at the latest 2.4.2p2 sources, and I can't find any place where it still might be doing this, but I don't see anything in the ChangeLog indicating such a change, nor has the docs/WISHLIST changed. Before I go to the trouble to build and install 2.4.2p2, can someone tell me if this has really changed or not? Thanks, David -- David E. Bernholdt | Email: [EMAIL PROTECTED] Oak Ridge National Laboratory| Phone: +1 (865) 574 3147 http://www.csm.ornl.gov/~bernhold/ | Fax: +1 (865) 574 0680
[Fwd: amdump taking over a day to complete]
Erick Bodine wrote: > The slow client is a Toshiba Magna 3000 with a single 18-GB SCSI disk. I am backing >up another partition on the same disk on the same host at the same time with no >problems. I do not see any disk or network errors on either the client or the tape >backup > server. > > Erick Bodine > Xpedite Systems Inc > Boulder CO > > Joshua Baker-LePain wrote: > > > On Thu, 21 Jun 2001 at 12:51pm, Erick Bodine wrote > > > > > I just added another partition to be backed up by amanda to my > > > configuration and it is taking over a day to back it up to tape. I know > > > amanda is still running because I do successive amstatus' [config] and > > > the percentage goes up incrementally. Do I need to change the dump > > > type to "compress server fast"? What is going on that it would take > > > this long to backup > > > > What type of box is the slow client? What type of disks? How is the > > networking set up? Are there any disk or network errors logged? Do you have > > a duplex mismatch anywhere? > > > > -- > > Joshua Baker-LePain > > Department of Biomedical Engineering > > Duke University
Re: Confused about multiple configurations
>Does this mean having multiple amanda.conf files; one for my "DailySet" >backups and another for my "ArchiveSet" backups, along with duplicate >directory structures for each? ... That's the basic idea. Much of the information between the two configurations can be shared. For amanda.conf, "includefile" is a big help. For instance, my tapetype entries are stored in a file that all the amanda.conf files bring in like this: includefile "../cf/tapetype" The disklist file can be symlinked between the two so any changes you make to the DailySet setup will automatically be seen in ArchiveSet. The rest of the areas (logfile, infofile and indexdir) need to be kept separate. You may or may not be able to share the holding disk (depends on whether DailySet and ArchiveSet will run at the same time). You can either have a slightly different dumptype (with dumpcycle 0) in ArchiveSet to do the level 0's, or you could do an "amadmin ArchiveSet force" before the run. John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
Re: Confused about multiple configurations
On Thu, 21 Jun 2001 at 2:58pm, Bill Delphenich wrote > Does this mean having multiple amanda.conf files; one for my "DailySet" > backups and another for my "ArchiveSet" backups, along with duplicate > directory structures for each? It doesn't look like you can define more > than one configuration per "amanda.conf" file. You're on exactly the right track. You *can* play tricks like having 'disklist' in your ArchiveSet etc directory be a link to the disklist in the DailySet etc directory. But, in my ArchiveSet disklist, I sometimes like to add disks that I don't hit daily (e.g. data disks I want a snapshot of), so I don't do that. > Is this documented somewhere that I haven't found yet? Is this a > relatively easy thing to do, or a difficult thing to do? It's pretty easy, and, like I said, you're on the right track. A couple of things to watch for: 1) Make sure that you set "record no" in all the dumptypes in your ArchiveSet config, so that it won't interfere with the DailySet schedule. 2) Make sure that the two configs don't run at the same time. You can't have two instances of amandad on a client at a time. 3) Set the ArchiveSet dumpcyle to 0 and tapecyle to something big. 4) Make sure that everything will fit on one tape or you have a changer defined. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Confused about multiple configurations
I am running a "DailySet" configuration five days a week as is common with AMANDA. Now I want to set up periodic manual full backups for archiving purposes. I don't want these to be scheduled. I just want to throw a tape in once a month or so and get a full-backup to store off-site. The FAQ says to set up multiple configurations to do this, but I'm not clear on how to do this. Does this mean having multiple amanda.conf files; one for my "DailySet" backups and another for my "ArchiveSet" backups, along with duplicate directory structures for each? It doesn't look like you can define more than one configuration per "amanda.conf" file. Is this documented somewhere that I haven't found yet? Is this a relatively easy thing to do, or a difficult thing to do? Thanks for any assistance.
Re: writing file: Input/output error [out of tape]
>I'm getting the following error ... That error comes from the operating system, not Amanda. Amanda is just reporting to you what the OS told it. You might check your system logs to see if any more detail is being reported, but even if it is it probably just translates to "media error" (one of my favorite "pass the buck" error codes :-). >... i use 40Gb DLT's ... Are you using a DLT8000 with DLT-IV tapes? That would be rated at 40 GBytes without hardware compression. Or are you using a DLT4000 with DLT-IV tapes? Those are rated at 20 GBytes without hardware compression and the only way you'd get 40 GBytes is with 50% compression, which depends a lot on the type of data you have, but my site certainly never get anywhere near that. Are you using hardware compression? If so, do you have software compression turned off? >... if i read right, Amanda >is complaining of running out of tape after less than 40Gb - the place where >the dump fails changes each day (after 2 Gb, 16 Gb, 20 Gb, 7 Gb, ...) The important line to look for is this one: taper: tape DAILY15 kb 16747360 fm 94 writing file: Input/output error This says taper got the error at ~16 GBytes. >I tried to use other tapes and another streamer ... Have you checked to make sure the bus is terminated properly (not too much, not too little, but just right :-)? You might check or swap out the cable, or at least make sure it is seated correctly. It's surprising how often that causes trouble. Next would be isolating the tape device so nothing else is on the bus. Have you tried cleaning the drive (even if it doesn't say it wants to be cleaned)? Are your tapes stored in a good environment (temperature, humidity)? Are your operators careful handling them? DLT media is particularly sensitive to rough treatment. A box of our tapes was dropped a couple of feet and every tape began to fail much like you're seeing. >Francis 'Dexter' Gois John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
amdump taking over a day to complete
I just added another partition to be backed up by amanda to my configuration and it is
taking over a day to back it up to tape. I know amanda is still running because I do
successive amstatus' [config] and the percentage goes up incrementally. Do I need
to change the dump type to "compress server fast"? What is going on that it would
take this long to backup
# Output from amstatus [config]
Using /var/log/amanda/RedBackUp1/amdump from Tue Jun 19 20:01:00 MDT 2001
137.236.93.118:/part2/home/nolgc 1 448k finished (20:02:37)
137.236.93.220:sda6 1 32k finished (20:02:35)
137.236.93.220:sda9 1 128k finished (20:04:05)
137.236.93.223:sda5 0 4775750k dumping 343104k ( 7.18%)
(1+11:36:50)
137.236.93.223:sda7 1 32k finished (20:02:38)
SUMMARY part real estimated
size size
partition : 5
estimated : 54776581k
failed : 0 0k ( 0.00%)
wait for dumping: 0 0k ( 0.00%)
dumping to tape : 0 0k ( 0.00%)
dumping : 1 343104k 4775750k ( 7.18%) ( 7.18%)
dumped : 4 640k 831k ( 77.02%) ( 0.01%)
wait for writing: 00k0k ( 0.00%) ( 0.00%)
writing to tape : 00k0k ( 0.00%) ( 0.00%)
failed to tape : 00k0k ( 0.00%) ( 0.00%)
taped : 4 640k 831k ( 77.02%) ( 0.01%)
3 dumpers idle : not-idle
taper idle
network free kps: 1970
holding space :96384k ( 1.88%)
dumper0 busy : 0:01:18 ( 0.06%)
dumper1 busy : 0:00:03 ( 0.00%)
dumper2 busy : 0:00:00 ( 0.00%)
taper busy : 0:00:00 ( 0.00%)
0 dumpers busy : 0:00:12 ( 0.01%) start-wait: 0:00:12 (100.00%)
1 dumper busy : 3:34:43 ( 9.04%)not-idle: 3:34:42 ( 99.99%)
2 dumpers busy : 3:38:18 ( 9.19%)not-idle: 3:38:16 ( 99.99%)
start-wait: 0:00:01 ( 0.01%)
3 dumpers busy : 3:35:57 ( 9.10%)not-idle: 3:35:56 ( 99.99%)
4 dumpers busy : 3:35:33 ( 9.08%)not-idle: 3:35:33 (100.00%)
5 dumpers busy : 3:36:03 ( 9.10%)not-idle: 3:36:03 (100.00%)
6 dumpers busy : 3:35:29 ( 9.08%)not-idle: 3:35:29 (100.00%)
7 dumpers busy : 3:35:47 ( 9.09%)not-idle: 3:35:47 (100.00%)
8 dumpers busy : 3:35:23 ( 9.07%)not-idle: 3:35:23 (100.00%)
9 dumpers busy : 3:35:33 ( 9.08%)not-idle: 3:35:33 (100.00%)
10 dumpers busy : 3:35:33 ( 9.08%)not-idle: 3:35:33 (100.00%)
11 dumpers busy : 3:35:40 ( 9.08%)not-idle: 3:35:40 (100.00%)
12 dumpers busy : 0:00:00 ( 0.00%)
# The current dumptype I am using.
define dumptype comp-high {
global
comment "very important partitions on fast machines"
compress client best
priority high
}
Erick Bodine
Xpedite Systems Inc
Boulder CO
FW: RedHat 7.1 & amrecover
We are running under RH7.1 having installed it using the RPM which I believe is on disk 2. I know it is recommended not to use the RPM but we had no problem with it. The only problem we had is in running amrecover, it complains that it cannot find the mount point. It then kicks you into the amrecover console where you signify the mount point with setdisk /. Then it works fine. Also note that when executing amrecover it complains about resolving the host name as it tries variations of your host until it reaches the name the server is set up with. Larry S. Brown MCSE President/CEO Dimension Networks, Inc. Member ICCA (727) 723-8388 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of [EMAIL PROTECTED] Sent: Thursday, June 21, 2001 11:25 AM To: [EMAIL PROTECTED] Subject: RedHat 7.1 & amrecover Still having problems with amrecover on RedHat 7.1 with amanda-2.4.2p2. I am wondering if my problem is with my configuration. I am assuming someone has run amrecover successfully on Redhat 7.1, so if you have let me know. Otherwise, can someone else try running amrecover on RedHat 7.1 if you have it installed? I thought my problem was with using localhost, so I went to using the actual hostname and amrecover does not work with it either. I have been able to use amdump, amrestore, amcheck, amverify, amadmin utilities successfully, just not amrecover. Another question, besides amdump creating the index file and amrecover using the index file, are there any other amanda utilities that use the index file? BTW John, the last note I sent, was sent too soon... I am reinstalling RedHat & amanda-2.4.2p2 to see if that makes a difference. I am also going to try SuSe with amanda-2.4.2p2 and see if that is any better. Thanks, Dan
Re: amidxtaped problem
>... By the way, since we're using an Exabyte 220 with 2 drives, >if we specify /dev/tape1, should we be able to do restores whilst backups >are running? Yes. >Carey John R. Jackson, Technical Software Specialist, [EMAIL PROTECTED]
New Admin, New Amanda User
Hello, I am a relatively new linux admin placed in a position to create new backups and backup procedures. We have decided to use Amanda as our backup utility, and I have a few questions: While it may seem silly, it is however a fact that I understand the concepts of how Amanda works and what takes place when it is working properly, however, I simply don't know how to get started. I have Amanda installed in a test environment, one server, one client. My problem is, even tho I know what's going on with it, I don't know how to begin. Which command do I run first? Which .conf file to edit? amanda.conf or create my own? The documentation is very good at explaining what Amanda CAN do, but not HOW TO do it. Something all unix beginners are aware of, advanced nix admin's aren't the best in the world at documenting their work. I would like someone to mentor me through the process and let me ask a few questions to get started. John Holstein Unix System Administrator CNpapers www.cnpapers.com
RedHat 7.1 & amrecover
Still having problems with amrecover on RedHat 7.1 with amanda-2.4.2p2. I am wondering if my problem is with my configuration. I am assuming someone has run amrecover successfully on Redhat 7.1, so if you have let me know. Otherwise, can someone else try running amrecover on RedHat 7.1 if you have it installed? I thought my problem was with using localhost, so I went to using the actual hostname and amrecover does not work with it either. I have been able to use amdump, amrestore, amcheck, amverify, amadmin utilities successfully, just not amrecover. Another question, besides amdump creating the index file and amrecover using the index file, are there any other amanda utilities that use the index file? BTW John, the last note I sent, was sent too soon... I am reinstalling RedHat & amanda-2.4.2p2 to see if that makes a difference. I am also going to try SuSe with amanda-2.4.2p2 and see if that is any better. Thanks, Dan
RE: amidxtaped problem
> > >We can't seem to get amrecover to work. We have to manually load the > >correct tape and then run amrecover, because if the correct tape > is not in, > >it's not able to access the tape device. > > Amrecover (et al) do not use the Amanda tape changer software (yet). > So, yes, you have to do the mounts by hand. > I see. Thanks. > > What did you pass amrecover for the -d option? You should do something > like "-d /dev/tape0". If you didn't pass a -d option, amrecover is > getting the default tape device from what you set with ./configure, > which should have been --with-tape-device=/dev/tape0. I'll try that. By the way, since we're using an Exabyte 220 with 2 drives, if we specify /dev/tape1, should we be able to do restores whilst backups are running? Thanks, Carey
time_t?
One of my filesystems failed with the following error: /-- arsd /net/data3 lev 0 FAILED [/tmpbin/tar returned 2] sendbackup: start [arsd:/net/data3 level 0] sendbackup: info BACKUP=/tmpbin/tar sendbackup: info RECOVER_CMD=/usr/local/bin/gzip -dc |/tmpbin/tar -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end ? gtar: time_t value 18446744071814433671 too large (max=68719476735) ? gtar: time_t value 18446744071814433668 too large (max=68719476735) | Total bytes written: 617922560 ? gtar: Error exit delayed from previous errors sendbackup: error [/tmpbin/tar returned 2] The tar version is 1.13 (tar --version), from sunfreeware.com. I tried to compile 1.13.19 from the sources, but it would always hang up at a certain point (I determined where it would fail, but not why). If someone could give me the exact version that is guaranteed to work (this is on a Solaris 8 machine) I would greatly appreciate it. Thanks Lance
Re: overwrite full with incremental?
On Jun 21, 2001, Pamela Miller <[EMAIL PROTECTED]> wrote:
> 0 20010615 DailySet112 3 1524850 1524864 788
> 3 20010621 DailySet112 3 252040 252064 218
You asked for one full backup per week (dumpcycle), and told Amanda to
use only 6 tapes (tapecycle), but it appears that you didn't run
Amanda 5 times during the week (runspercycle), which might explain why
it wouldn't have found enough space in the tape to run another full
backup of the disk, especially given that it wasn't due yet (since a
week hadn't passed).
It's strongly recommended that you have additional tapes ((2 *
runspercycle + 1) * runtapes) in the tapecycle in case a full backup
fails: then you can always go back to the other, without the risk of
having had it just overwritten.
Amanda probably made a lot of noise in the mail backups warning that
the last full backup of that disk was about to be overwritten, and
finally overwritten. This is a strong indication that your tapecycle
is too short, or your dumpcycle is too long, or that you aren't
following the runspercycle plan.
--
Alexandre Oliva Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicampoliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist*Please* write to mailing lists, not to me
Re: amrecover problem
Am Mittwoch, 20. Juni 2001 22:17 schrieben Sie: Hi John! > >For security: My NoRewinding device ist /dev/nst0. > >Shall i take this device for my changer? > It should be the tape device name. Whether it is also your changer > device name depends on your hardware. > >When i take > >amrecover -d /dev/nst0 or > >amrecover -d /dev/st0 > >and i become the same error ... > Using /dev/nst0 should have worked. I'm not sure I understood your > last sentence. Did "-d /dev/nst0" fail? What messages did you get? > In particular (if it failed), what was in /tmp/amanda/amidxtaped*debug? > >sorry, i forgot to post a message called: > > WARNING: not on root of selected filesystem, check man-page! > >I think the $CWD mount point is not set correct. > You didn't post the important message that comes out right before this: > $CWD '/some/path' is on disk '/some/disk' mounted at /some/mount/point'. > Seeing that would help understand the problem. That's fine. > >When i go to /drives/drive2 and start amrecover, the Warning Message don't > >come and restore works (i hope so) > But the files replace the originals, right? right, but before it comes a warning message that the files exists. > >When i go to /drives/drive2/Online-Marketing/customer/April the warning > >message is coming and the restore failed. > What do you mean by "the restore failed"? I don't no, amrestore is quiting with no comment. > >Where must i go to use the right $CWD mount point? > When I do restores, I create a directory at the top of the file system > that was backed up, e.g. "/drives/drive2/.bkrest", then "cd" to there > and start amrecover. > > If you cd to /drives/drive2/Online-Marketing/customer/April and run > amrecover from there, files will be restored relative to that point. > For instance, restoring Online-Marketing/customer/April/JJ would come > back as: > /drives/drive2/Online-Marketing/customer/April/Online-Marketing/customer/Ap >ril/JJ > > which is probably not what you want (although you could rename it, > of course). That's why I create the temp area at the top of the mount > point so there is only one extra directory layer involved. well, i made a directory so you've written, i used the /dev/nst0 and the recovery is working fine. Thank you very much for your Info. That helps! Thanks a lot. Bye Juergen -- Dies ist eine Microsoft freie Mail!
overwrite full with incremental?
Hi, I've just started running Amanda. But, it looks like Amanda has overwritten a full dump with an incremental, which seems wrong? For /export/home on host stheno why is Daily112 listed as having both a full (lv0) and incremental (lv3)? If I start an amrecover, I am only prompted for Daily112Is Amanda doing the right thing or is my config wrong? Thanks. bash-2.03$ ./amadmin daily info stheno /export/home/ Current info for stheno /export/home/: Stats: dump rates (kps), Full: 1935.0, 1359.0, -1.0 Incremental: 1156.0, 968.0, 1224.0 compressed size, Full: -100.0%,-100.0%,-100.0% Incremental: -100.0%,-100.0%,-100.0% Dumps: lev datestmp tape file origK compK secs 0 20010615 DailySet112 3 1524850 1524864 788 1 20010616 DailySet115 3 389880 389888 208 2 20010618 DailySet113 3 590580 590592 313 3 20010621 DailySet112 3 252040 252064 218 bash-2.03$ ./amadmin daily find stheno /export/home/ Scanning /home/pam/amanda/dumps... date host disk lv tape or file file status 2001-06-16 stheno /export/home/ 1 DailySet115 3 OK 2001-06-18 stheno /export/home/ 2 DailySet113 3 OK 2001-06-19 stheno /export/home/ 3 DailySet116 3 OK 2001-06-20 stheno /export/home/ 3 DailySet111 3 OK 2001-06-21 stheno /export/home/ 3 DailySet112 3 OK and amrecover history shows no full dumps? amrecover> history 200- Dump history for config "daily" host "stheno" disk "/export/home/" 201- 2001-06-21 3 DailySet112 3 201- 2001-06-20 3 DailySet111 3 201- 2001-06-19 3 DailySet116 3 201- 2001-06-18 2 DailySet113 3 201- 2001-06-16 1 DailySet115 3 200 Dump history for config "daily" host "stheno" disk "/export/home/" >From my amanda.config: dumpcycle 7 runspercycle 5 tapecycle 6 tapes /etc/amandates get written to at the time of the backup and contains (on stheno): /export 0 992014907 /export/home/ 0 992587772 /export/home/ 1 992690883 /export/home/ 2 992880883 /export/home/ 3 993108916 I have taken most of the default settings from the example config file and not set anything special like, the skip-full or skip-inc variables. My dump type in user-tar, I think I just took it from the example config file. Pam
writing file: Input/output error [out of tape]
Hi,
I'm getting the following error about each day while running my backup. It
seems very strange to me since i use 40Gb DLT's and if i read right, Amanda
is complaining of running out of tape after less than 40Gb - the place where
the dump fails changes each day (after 2 Gb, 16 Gb, 20 Gb, 7 Gb, ...)
I tried to use other tapes and another streamer, i get exactly the same
result. I also made a tar directly to the streamer. Maybe i was lucky, but i
didn't get any error - that makes me think the problem is not hardware but
amanda config.
My DLT are configured like this in amanda.conf :
tapetype DLT# what kind of tape it is (see tapetypes below)
define tapetype DLT {
comment "DLT tape drives"
length 38500 mbytes
filemark 2000 kbytes
speed 1000 kbytes
}
Has anybody an idea of what could be the problem, how i can solve it or other
infos ?
Thanks a lot.
Regards,
Dexter
Error messages in the mail report :
---
*** A TAPE ERROR OCCURRED: [[writing file: Input/output error]].
*** PERFORMED ALL DUMPS TO HOLDING DISK.
THESE DUMPS WERE TO DISK. Flush them onto a new tape.
Tonight's dumps should go onto 1 tape: a new tape.
NOTES:
driver: WARNING: /var/spool/backup/holdingdisk: 11534336 KB requested, but
only 6637460 KB available.
driver: WARNING: /var/local/space/holdingdisk: 31457280 KB requested, but
only 19919464 KB available.
planner: Last full dump of oracle.freegates.net:/home on tape DAILY02
overwritten in 4 runs.
planner: Incremental of dbstat:/var bumped to level 2.
planner: Incremental of testradius3:/var bumped to level 3.
planner: Dump too big for tape: full dump of beta:/var delayed.
taper: tape DAILY15 kb 16747360 fm 94 writing file: Input/output error
end of file /var/log/amanda/Normal/log :
INFO taper tape DAILY15 kb 16747360 fm 94 writing file: Input/output error
FAIL taper oracle.freegates.net /home 0 [out of tape]
ERROR taper no-tape [[writing file: Input/output error]]
FAIL dumper oracle.freegates.net /home 0 ["data write: Broken pipe"]
sendbackup: start [oracle.freegates.net:/home level 0]
sendbackup: info BACKUP=/usr/bin/gtar
sendbackup: info RECOVER_CMD=/usr/bin/gtar -f... -
sendbackup: info end
FAIL driver oracle.freegates.net /home 0 [dump to tape failed]
FATAL taper syncpipe_get: w: unexpected EOF
end of file /var/log/amanda/Normal/amdump :
---
taper: writing end marker. [DAILY15 ERR kb 16747360 fm 94]
driver: result time 8096.332 from dumper0: FAILED 01-00188 ["data write:
Broken pipe"]
driver: result time 8096.332 from taper: TAPE-ERROR 00-00187 [writing file:
Input/output error]
dumper: kill index command
driver: QUITTING time 8096.389 telling children to quit
driver: send-cmd time 8096.389 to taper: QUIT
driver: send-cmd time 8096.389 to dumper0: QUIT
driver: send-cmd time 8096.389 to dumper1: QUIT
driver: send-cmd time 8096.389 to dumper2: QUIT
driver: send-cmd time 8096.389 to dumper3: QUIT
driver: FINISHED time 8096.459
amdump: end at Wed Jun 20 18:43:44 CEST 2001
--
Knowledge-sharing and open-source content : another way to gain eternity.
Francis 'Dexter' Gois - [EMAIL PROTECTED]
System Administrator - Tiscali Belgium NV/SA
phone: +3224000839- fax : +3224000899
Re: New user, a few questions to get it going
On Thu, Jun 21, 2001 at 05:24:15AM +0700, Olivier Nicole ([EMAIL PROTECTED]) wrote: > JObst, > > A couple of answers: > > 1) it seems that for reason amanda did install itself with owner root > instead of bin. When you execute it with the wrong group as the wrong user thi does happen. Its fixed now, I added the wrong group. thanks jobst -- You win some, you loose some. |__, Jobst Schmalenbach, [EMAIL PROTECTED], Technical Director| | _ _.--'-n_/ Barrett Consulting Group P/L & The Meditation Room P/L | |-(_)--(_)= +61 3 9532 7677, POBox 277, Caulfield South, 3162, Australia|
Re: New user, a few questions to get it going
On Wed, Jun 20, 2001 at 09:34:31AM -0400, Joshua Baker-LePain ([EMAIL PROTECTED]) wrote: > On Wed, 20 Jun 2001 at 10:01pm, Jobst Schmalenbach wrote > That's normal. Understood, its a -w flag. > > > permissions on amcheck: > > > > 452] [piquet:root] /usr/local/sbin #ROOT# >ls -la > > total 2896 > > drwxr-xr-x2 root root 4096 Jun 20 21:14 . > > drwxr-xr-x 20 root root 4096 Jun 20 10:14 .. > > -rwsr-x---1 root disk 393508 Jun 20 21:14 amcheck > > I'm guessing that 'bin' isn't in the 'disk' group... True, you wont believe how many times I looked at it and did not see it! it was the OTHER way around. [snip] > > WARNING: localhost: selfcheck request timed out. Host down? > > > > Even if I change it to the real name/ipaddress it doesnt work. > > I filled the ipaddress into the xinted.conf file and added all > > the stuff in there! > > You shouldn't need to make any changes to xinetd.conf. You need the Actaully where you do it (either in xinted.conf or add another file ) does not really matter . > proper files in /etc/xinetd.d and to allow the services in > /etc/hosts.allow. > > > I can list the socket: > > [494] [piquet:root] /tmp/amanda #ROOT# >netstat -a | grep -i amanda > > tcp0 0 *:amandaidx *:* LISTEN > > Your amanda service isn't running. amandaidx is for listening for > amrecover requests from other hosts. Thats correct as I know now. The permission was the problem. [snip] > -- > Joshua Baker-LePain > Department of Biomedical Engineering > Duke University Thanks for your help, its fixed. jobst -- Keyboard not found - please clean up desktop! |__, Jobst Schmalenbach, [EMAIL PROTECTED], Technical Director| | _ _.--'-n_/ Barrett Consulting Group P/L & The Meditation Room P/L | |-(_)--(_)= +61 3 9532 7677, POBox 277, Caulfield South, 3162, Australia|
interesting auction
Just peeked at egghead.com's auctions. They have an HP SureStore DLT Tape Library available. 40 tape / 6 drive capacity (none supplied). Item is listed as new, est. list price $40K (1 bid at $1.4K) Not something I need for my home lan, but someone on the list might be interested in 3.5 TB capacity :) -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing [EMAIL PROTECTED] 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: New user, a few questions to get it going
On Wed, Jun 20, 2001 at 02:58:00PM -0500, John R. Jackson ([EMAIL PROTECTED]) wrote: > > I am new to amanda ... [snip] > As Joshua Baker-LePain wrote, it looks like most of your problems are > related to not putting "bin" in group "disk". Thats it! [snip] > The write test is "destructive", i.e. whatever backup images were on the > tape will no longer be accessible. Most people like to play it safe and > not lose them "just in case". If you really want to enable the test to > make sure the tape is writable, add the "-w" flag to "amcheck". understood. > > >... In disklist to get this running (finetuning later) I have: > >localhost / root-tar > > Please don't use "localhost". Use the fully qualified domain name. > "localhost" has "issues" that will eventually bite you. OK. [snip] > And one deletion. Don't put "server_args amandad". That's a I know, xinted is soo much better thanks for your help! jobst -- A loving atmosphere in your home is the foundation for your life. |__, Jobst Schmalenbach, [EMAIL PROTECTED], Technical Director| | _ _.--'-n_/ Barrett Consulting Group P/L & The Meditation Room P/L | |-(_)--(_)= +61 3 9532 7677, POBox 277, Caulfield South, 3162, Australia|
