Re: Amanda I/O error!
On Mon, 13 May 2002, Joshua Baker-LePain wrote: > On Mon, 13 May 2002 at 6:23pm, Alexander Belik wrote > > > I have Input/output error. But I have cleaned my device. > Dirty tapes aren't the only source of I/O errors. Look in your system > logs for what type of I/O error the kernel reported. I found errors but ! What it meen? What can I do? -- May 15 03:30:21 auth kernel: st0: Error with sense data: [valid=0] Info fld=0x0, Current st09:00: sense key Unit Attention May 15 03:30:21 auth kernel: st0: Error with sense data: [valid=0] Info fld=0x0, Current st09:00: sense key Unit Attention May 15 03:30:21 auth kernel: Additional sense indicates Power on, reset, or bus device reset occurred May 15 03:30:21 auth kernel: Additional sense indicates Power on, reset, or bus device reset occurred -- [EMAIL PROTECTED]ICQ: 41776461 2:465/207@Fidonet Alexander Belik http://www.vnet.dn.ua[EMAIL PROTECTED]
RE: Following symlinks
Your tip certainly helped me, Michael, and actually your example _made_
me change to /backup/ instead og ./backup, since the old directory path
was painfully long...
Thank you for helping me.
;-Peter
-Original Message-
From: Michael Richardson [mailto:[EMAIL PROTECTED]]
Sent: 15. maj 2002 03:12
To: Bernhard R. Erdmann
Cc: Peter Normann; [EMAIL PROTECTED]
Subject: Re: Following symlinks
> "Bernhard" == Bernhard R Erdmann <[EMAIL PROTECTED]> writes:
>> reference the files as:
>>
>> /.backup/whatever/.
>>
>> The trailing /. will force the symlink to get dereferenced.
Bernhard> Didn't you mean ./backup/whatever/.?
I read that he had created the symlinks in /.backup/, but ...
whatever.
] ON HUMILITY: to err is human. To moo, bovine. |
firewalls [
] Michael Richardson, Sandelman Software Works, Ottawa, ON|net
architect[
] [EMAIL PROTECTED] http://www.sandelman.ottawa.on.ca/ |device
driver[ ] panic("Just another NetBSD/notebook using, kernel hacking,
security guy"); [
Re: Following symlinks
> "Bernhard" == Bernhard R Erdmann <[EMAIL PROTECTED]> writes:
>> reference the files as:
>>
>> /.backup/whatever/.
>>
>> The trailing /. will force the symlink to get dereferenced.
Bernhard> Didn't you mean ./backup/whatever/.?
I read that he had created the symlinks in /.backup/, but ... whatever.
] ON HUMILITY: to err is human. To moo, bovine. | firewalls [
] Michael Richardson, Sandelman Software Works, Ottawa, ON|net architect[
] [EMAIL PROTECTED] http://www.sandelman.ottawa.on.ca/ |device driver[
] panic("Just another NetBSD/notebook using, kernel hacking, security guy"); [
Re: amanda - my way ?
> But it was a hard fight to convince them to do backups parallel > to their raid-5 array. > Their argument: why backup, we have raid rm -rf /
Re: Question in case of disaster
> I run amanda-2.4.2p1-2. Supose an intruder break somehow the system and > perform a "rm -rf /". From this point, after a fresh reinstall of the > system, how I can restore from tape, if no backup database is not > availale anymore? I use rsync via ssh after the backup jobs to copy Amanda's index to another box.
Re: Following symlinks
> reference the files as: > > /.backup/whatever/. > > The trailing /. will force the symlink to get dereferenced. Didn't you mean ./backup/whatever/.?
Re: Question in case of disaster
On Tue, 14 May 2002, Deb Baddorf wrote: > I could be wrong here, but: > I think the part they want you to print is only an OVERVIEW -- > which disk entry and what number file it is on the tape. > NOT a listing of individual files. That was what I meant anyway. The important part for me is the file number, ie where to move with 'mt' in case a manual restore is required. The professional look of the tape cases is a nice bonus. -- Ulrik Sandberg
Re: Amanda dump level policy
> So if I want to persuade amanda to use more tape I should shorten the dump > cycle, correct ? But I have a DDS-3 with native capacity of 12 GB and no > changer. What happens if in a dump cycle there's more to be backed up than > can fit on the tape e.g. if a lot happens to change on a couple of > filesystems and level 1s get large ? What falls off the edge of the world ? It depends on your value of tapesize and the size of your holding disk. If the backup volume is larger than a single tape, but below holding disk size, the rest will stay on disk waiting to be amflushed. If space is really a shortage, Amanda will decide to leave off subsequent full dumps to do some more incrementals. These are prioritized according to your disklist entries and their configured priority.
Re: Question in case of disaster
I could be wrong here, but: I think the part they want you to print is only an OVERVIEW -- which disk entry and what number file it is on the tape. NOT a listing of individual files. If your whole disk is bad, you don't need an index very much -- you won't want to restore individual files. You want to restore the whole disk. And therefore the short list (just the disk names) is useful. If I'm hearing both sides aright .But I could be wrong! Deb Baddorf At 06:24 PM 5/14/2002 +0300, you wrote: >On Mon, 13 May 2002, Anthony A. D. Talltree wrote: > > > There's a lot to be said for printing tape labels or case inserts that > > document the contents of each tape -- or for printing each day's results > > and keeping them in a binder. > >In my case this cannot be done: I have a mission critical server, with >thousand of small files and there is an ocean between my location and >server location. In case of a disaster I have to be able to restore a huge >directory tree with more than 10,000 files within minutes or hours at >most. With a paper list and tapes that I have to get a visa and fly a day >in order to touch them this is not an option. --- Deb Baddorf [EMAIL PROTECTED] 840-2289 "You can't help getting older, but you don't have to get old." - George Burns <
Re: Question in case of disaster
On Tue, 14 May 2002, Anthony A. D. Talltree wrote: > >In case of a disaster I have to be able to restore a huge > >directory tree with more than 10,000 files within minutes or hours at > >most. With a paper list and tapes that I have to get a visa and fly a day > >in order to touch them this is not an option. > > Is there a compelling reason why you can't print them out where you are? Yes, there are: - I cannot afford a printer and so much paper - I have slow connection and fetching data will takes hours And most important: in case of a disaster, I'll be lost in a sea of paper trying to restore +10,000 files in few hours. I'm stop writing on this thread, since it does not sense to continue form this point. Ave, Radu -- Radu Filip Network Administrator @ Technical University of Iasi [EMAIL PROTECTED] Information Technology and Communication Center http://socrate.tuiasi.ro/ [EMAIL PROTECTED] | http://ccti.tuiasi.ro/
Re: Tapecycle question
On Tue, May 14, 2002 at 11:37:51AM -0500, Dave Sherohman wrote: > > tapecycle parameter do ? If there are less than tapecycle tape in the > > tapelist will it force amdump to ask for a new one ? > > Yep. A tape must be at least tapes old before amanda will > agree to reuse it. So for you, with tapecycle set to 14, amanda will > refuse to overwrite the last 14 tapes that have been used. Why then does my tapelist have reuse at the end of every single tape's entry ? > I always amflush everything to a single tape. AFAIK, the only > disadvantage to doing it that way is that, if that one tape goes bad, > you're losing more backups. (And, similarly, if you do what you're > talking about and your holding drive dies on Sunday night...) OK - both caveats understood. If there are people working they can change the tape. Otherwise, it's just some extra backups and if a tape or the holding disk goes bad, it's not a big loss. Thanks, Niall O Broin
Re: sorry to bother with that
are you able to control the library with chg-scsi? i'm running amanda on solaris8 with an exabyte ez17 library and i had to use chg-zd-mtx. the host down error usually means the server can't talk to the client, can you ping it? stupid question i know but you have to ask. is the amanda client software running, netstat -a |grep am should tell you? you said that your using NIS, personally i'd still add the services to /etc/services if you only have it in NIS. in amanda's $HOME do you have the .amandahosts file? if should contain lines like: amandaserver amanda amandaserver.domain.com amanda without that you'll never get logged in. take a look at the /tmp/amanda directory, if it doesn't exist create it. you should be able to get alot of good debug info from there. -- Darin Perusich Unix Systems Administrator Cognigen Corp. [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: > > Hi guys, > Sorry I try to make amanda work on my server (solaris 8) and after reading all > paper and try and try, not working > Can yny body give some help. > First with the amanda.conf, I try one and put it under /opt/etc/amanda/csd (csd > is suppose to be my configuration directory). > For now Id like to try a backup every week, and later I will have to change > that, some server need every day some not. > So here is my amanda.conf, and my disklist (where I suppose to put this one ), > and the chg-scsi > My tape changer is an HP SureStore 12000e (6 time 24gb), device /dev/rmt/2 > > (See attached file: disklist)(See attached file: chg-scsi.conf)(See attached > file: amanda.conf) > > Here is my amcheck > orion:/opt/etc/amanda#amcheck csd > Amanda Tape Server Host Check > - > Holding disk /opt/etc/amanda/dump: 2457285 KB disk space available, that's > plenty > amcheck-server: could not get changer info: could not read result from > "/usr/local/libexec/chg-scsi" (got signal 11) > > Amanda Backup Client Hosts Check > > WARNING: orion: selfcheck request timed out. Host down? > WARNING: hyperion: selfcheck request timed out. Host down? > Client check: 2 hosts checked in 30.073 seconds, 2 problems found > > (brought to you by Amanda 2.4.2p2) > > > Why is my server down ? I use nnis and the services ar setup correctly. > Hyperion is not working since I didn't install the software yet, later, when I > understand what I'm doing on the server > > Thanks, > > Arno > > > ** > DISCLAIMER - E-MAIL > --- > The information contained in this E-Mail is intended for the named > recipient(s). It may contain certain privileged and confidential > information, or information which is otherwise protected from > disclosure. If you are not the intended recipient, you must not > copy,distribute or take any action in reliance on this information > ** > >
Re: Question in case of disaster
On Tue, 14 May 2002, Radu Filip wrote: - On Mon, 13 May 2002, Anthony A. D. Talltree wrote: - - > There's a lot to be said for printing tape labels or case inserts that - > document the contents of each tape -- or for printing each day's results - > and keeping them in a binder. - - In my case this cannot be done: I have a mission critical server, with - thousand of small files and there is an ocean between my location and - server location. In case of a disaster I have to be able to restore a huge - directory tree with more than 10,000 files within minutes or hours at - most. With a paper list and tapes that I have to get a visa and fly a day - in order to touch them this is not an option. - - As a workaround, I'm backing up all /var/lib/amanda/my_tapes on a diffrent - machine, but this is just an workaround. This is why I asked if baking up - index itself on the beginning of a tape so Amanda can use it as an - alternative it is planned as a future feature for Amanda, because in cases - when you need to restore everything in shortest time, this is really - usefull. What I am implementing here is a second Amanda tape server. After the backups are done on the primary, I run rsync to syncronoze all of the relavalent directories to the backup machine. -- -- Stephen Carville UNIX and Network Administrator DPSI (formerly Ace USA Flood Services) 310-342-3602 [EMAIL PROTECTED]
data timeouts...
I am having problems with timeouts on one of my file systems. I am running version 2.4.2p2 on Solaris 8 (Sparc). I am doing backup with GNUtar version 1.13.19. I am using tar to backup directories on a large raid (larger then my tape size). All the file systems execpt for one are working fine. This one file system is where we store mail so it has lots of small files. Here is what I am getting in my amanda report: nfs-1c-bvt /export/raid/mail lev 2 FAILED [data timeout] /-- nfs-1c-bvt /export/raid/mail lev 2 FAILED [data timeout] sendbackup: start [nfs-1c-bvtn:/export/raid/mail level 2] sendbackup: info BACKUP=/usr/local/bin/tar sendbackup: info RECOVER_CMD=/bin/gzip -dc |/usr/local/bin/tar -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end and the sendbackup debug file on the client: sendbackup: debug 1 pid 22417 ruid 105 euid 105 start time Tue May 14 08:00:31 2002 /private/amanda/libexec/sendbackup: version 2.4.2p2 sendbackup: got input request: GNUTAR /export/raid/mail 2 2002:5:11:10:31:12 OPTIONS |;bsd-auth;compress-fast;ind ex; parsed request as: program `GNUTAR' disk `/export/raid/mail' lev 2 since 2002:5:11:10:31:12 opt `|;bsd-auth;compress-fast;index;' sendbackup: try_socksize: send buffer size is 65536 sendbackup: stream_server: waiting for connection: 0.0.0.0.41164 sendbackup: stream_server: waiting for connection: 0.0.0.0.41165 sendbackup: stream_server: waiting for connection: 0.0.0.0.41166 waiting for connect on 41164, then 41165, then 41166 sendbackup: stream_accept: connection from 192.168.2.101.40571 sendbackup: stream_accept: connection from 192.168.2.101.40572 sendbackup: stream_accept: connection from 192.168.2.101.40573 got all connections sendbackup: spawning /bin/gzip in pipeline sendbackup: argument list: /bin/gzip --fast sendbackup-gnutar: pid 22418: /bin/gzip --fast sendbackup-gnutar: doing level 2 dump as listed-incremental from /private/amanda/var/amanda/gnutar-lists/nfs-1c-b vtn_export_raid_mail_1 to /private/amanda/var/amanda/gnutar-lists/nfs-1c-bvtn_export_raid_mail_2.new sendbackup-gnutar: doing level 2 dump from date: 2002-05-11 10:31:16 GMT sendbackup: started index creator: "/usr/local/bin/tar -tf - 2>/dev/null | sed -e 's/^\.//'" sendbackup: spawning /private/amanda/libexec/runtar in pipeline sendbackup: argument list: gtar --create --file - --directory /export/raid/mail --one-file-system --listed-increm ental /private/amanda/var/amanda/gnutar-lists/nfs-1c-bvtn_export_raid_mail_2.new --ignore-failed-read --totals . sendbackup-gnutar: /private/amanda/libexec/runtar: pid 22423 index tee cannot write [Broken pipe] sendbackup: pid 22419 finish time Tue May 14 09:02:46 2002 etimeout is set to 4 in amanda.conf. Any ideas? What should I be looking at? Thanks for your help, Luke * * Luke Miller Unix System Administrator * * Integra Telecom 503-748-4549 * *
Re: Tapecycle question
On Tue, May 14, 2002 at 09:59:45AM +0100, Niall O Broin wrote: > and asked for #15 for tonight. From what I've read before, amdump will use > the last tape in the tapelist marked reuse (incidentally, every tape in my > tapelist is marked reuse - is that correct ?) so what exactly does the > tapecycle parameter do ? If there are less than tapecycle tape in the > tapelist will it force amdump to ask for a new one ? Yep. A tape must be at least tapes old before amanda will agree to reuse it. So for you, with tapecycle set to 14, amanda will refuse to overwrite the last 14 tapes that have been used. > I've got a large holding disk so I may well change runspercycle to 14 and > run amdump every day and then run amflush on Monday in order to have as many > backups as possible (and people may sometimes do some work at weekends). If > I do that and the amount of data in the two holding directories on Monday is > small enough, is it OK to amflush ALL to one tape or am I better off > flushing each day to a separate tape, which would give the same nett result > as if the tapes had been put in the drive over the weekend ? I always amflush everything to a single tape. AFAIK, the only disadvantage to doing it that way is that, if that one tape goes bad, you're losing more backups. (And, similarly, if you do what you're talking about and your holding drive dies on Sunday night...)
Re: Getting old mails ?
On Tue, May 14, 2002 at 09:22:24AM +0100, Niall O Broin wrote: > Is it just me, or have other people on the list been getting old mails again > - I just got one from Michael Richard about problems with firewalls. I was > sure I'd read it before and sure enough - date in header was May 5. > Looking at the headers the problem (with this one anyway) appears to be on > the server hongkong.com. There have been a bunch of them. I've already informed [EMAIL PROTECTED] They tell me the problem is addressed (by refusing hongkong.com posts). -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Question in case of disaster
>In case of a disaster I have to be able to restore a huge >directory tree with more than 10,000 files within minutes or hours at >most. With a paper list and tapes that I have to get a visa and fly a day >in order to touch them this is not an option. Is there a compelling reason why you can't print them out where you are?
Re: forcing a degraded backup
On Tue, May 14, 2002 at 11:34:21AM -0400, Jean-Louis Martineau wrote: > It should work if you allow full dump in degraded mode by setting the > reserve parameter. It's the best solution if you have enough holding disk. I can't really allow full dumps in degraded mode because I have some very large filesystems (one of 60G, which hopefully will always compress enouth to fit on tape) so if there is too much full dumps to flush the tape may be too small. > > If you don't allow full dump in degraded mode, then I thing that setting > runspercycle=7 will give better result at balancing full dump. OK, I'll let it run a few days and see how it behave. Thanks for the advices ! -- Manuel Bouyer, LIP6, Universite Paris VI. [EMAIL PROTECTED] --
sorry to bother with that
Hi guys, Sorry I try to make amanda work on my server (solaris 8) and after reading all paper and try and try, not working Can yny body give some help. First with the amanda.conf, I try one and put it under /opt/etc/amanda/csd (csd is suppose to be my configuration directory). For now Id like to try a backup every week, and later I will have to change that, some server need every day some not. So here is my amanda.conf, and my disklist (where I suppose to put this one ), and the chg-scsi My tape changer is an HP SureStore 12000e (6 time 24gb), device /dev/rmt/2 (See attached file: disklist)(See attached file: chg-scsi.conf)(See attached file: amanda.conf) Here is my amcheck orion:/opt/etc/amanda#amcheck csd Amanda Tape Server Host Check - Holding disk /opt/etc/amanda/dump: 2457285 KB disk space available, that's plenty amcheck-server: could not get changer info: could not read result from "/usr/local/libexec/chg-scsi" (got signal 11) Amanda Backup Client Hosts Check WARNING: orion: selfcheck request timed out. Host down? WARNING: hyperion: selfcheck request timed out. Host down? Client check: 2 hosts checked in 30.073 seconds, 2 problems found (brought to you by Amanda 2.4.2p2) Why is my server down ? I use nnis and the services ar setup correctly. Hyperion is not working since I didn't install the software yet, later, when I understand what I'm doing on the server Thanks, Arno ** DISCLAIMER - E-MAIL --- The information contained in this E-Mail is intended for the named recipient(s). It may contain certain privileged and confidential information, or information which is otherwise protected from disclosure. If you are not the intended recipient, you must not copy,distribute or take any action in reliance on this information ** disklist Description: Binary data chg-scsi.conf Description: Binary data amanda.conf Description: Binary data
Re: Question in case of disaster
On Tue, 14 May 2002 at 6:24pm, Radu Filip wrote
> As a workaround, I'm backing up all /var/lib/amanda/my_tapes on a diffrent
> machine, but this is just an workaround. This is why I asked if baking up
> index itself on the beginning of a tape so Amanda can use it as an
> alternative it is planned as a future feature for Amanda, because in cases
> when you need to restore everything in shortest time, this is really
> usefull.
You currently have a couple options:
1) Back up the {log,index}dir to tape with amanda (with dumpcycle 0, so
you get a full snapshot every night). In case of a failure, install
amanda and use amrestore to find the {log,index}dir, restore that, and
then restore the VIF (very important filesystem).
2) Every night, as part of your backup script, tar up the {log,index}dir
and (s)cp it to another machine. In case the primary machine dies, you'll
still have your entire amanda history. for extra redundancy, back up that
directory on the other machine as well.
I currently do *both* these things, and in some senses consider it a
minimum safe practice.
--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University
Re: forcing a degraded backup
On Mon, May 13, 2002 at 04:37:35PM +0200, Manuel Bouyer wrote: > On Tue, May 07, 2002 at 07:52:10AM -0400, Jean-Louis Martineau wrote: > > > > That's BAD, you can't use the same logdir for two configurations, logfile > > > > will get erased. You will get the same problem with indexdir if you have > > > > a disk in both config. > > > > > > I don't understand why ... note that both configs won't run at the same day > > > (one day for backup to tape, next day a degraded backup). > > > When I use 'history' in amrecover I get all the backups in the list, > > > with the proper tapes. Also the log files are all there. > > > > It works now because it's a new config, After tapecycle days, one config > > will erase the log and index of the other config. > > Ha OK. Hum, I can see a problem even with one config then: > I think I should have dumpcycle=2 weeks and runspercycle=14, (one run per day) > but I'll have only have half the expected tapes (as one run out of 2 > will go to holding disk and will be flushed with the dump of the next run). > I'm not sure how amanda will handle it. I already planned to have 14 tapes > to have 2 cycles of tapes, so it should be OK. It should work if you allow full dump in degraded mode by setting the reserve parameter. It's the best solution if you have enough holding disk. If you don't allow full dump in degraded mode, then I thing that setting runspercycle=7 will give better result at balancing full dump. Jean-Louis > > > > > > > Why not use one config and edit your amanda.conf everyday, that could > > > > be easily done by cron. > > > > > > I though at this, but I prefer static config files. This avoids problems > > > in case the server goes down in the middle of a backup. > > > > I think that running a small sed script to patch a static config > > file template before every amdump run is easier than using two configs. > > OK, I'll look at this. > My problem with this approach is that if the server crashes in the middle of > a degraded dump, then the wrong config will be in place until the next > run, and amtape won't work (for use with amrecover for example). I agree this > would be very bad luck. > Also it has to be changed for amcheck too. > > -- > Manuel Bouyer, LIP6, Universite Paris VI. [EMAIL PROTECTED] > -- -- Jean-Louis Martineau email: [EMAIL PROTECTED] Departement IRO, Universite de Montreal C.P. 6128, Succ. CENTRE-VILLETel: (514) 343-6111 ext. 3529 Montreal, Canada, H3C 3J7Fax: (514) 343-5834
Re: Question in case of disaster
On Mon, 13 May 2002, Anthony A. D. Talltree wrote: > There's a lot to be said for printing tape labels or case inserts that > document the contents of each tape -- or for printing each day's results > and keeping them in a binder. In my case this cannot be done: I have a mission critical server, with thousand of small files and there is an ocean between my location and server location. In case of a disaster I have to be able to restore a huge directory tree with more than 10,000 files within minutes or hours at most. With a paper list and tapes that I have to get a visa and fly a day in order to touch them this is not an option. As a workaround, I'm backing up all /var/lib/amanda/my_tapes on a diffrent machine, but this is just an workaround. This is why I asked if baking up index itself on the beginning of a tape so Amanda can use it as an alternative it is planned as a future feature for Amanda, because in cases when you need to restore everything in shortest time, this is really usefull. Ave, Socrate -- Radu Filip Network Administrator @ Technical University of Iasi [EMAIL PROTECTED] Information Technology and Communication Center http://socrate.tuiasi.ro/ [EMAIL PROTECTED] | http://ccti.tuiasi.ro/
Getting old mails ?
Is it just me, or have other people on the list been getting old mails again - I just got one from Michael Richard about problems with firewalls. I was sure I'd read it before and sure enough - date in header was May 5. Looking at the headers the problem (with this one anyway) appears to be on the server hongkong.com. Regards, Niall O Broin
Tapecycle question
What exactly does tapecycle do ? I have in my config dumpcycle 2 weeks # the number of days in the normal dump cycle runspercycle 10 # the number of amdump runs in dumpcycle days # (2 weeks * 5 amdump runs per week -- just weekdays) tapecycle 14 tapes # the number of tapes in rotation I have 22 tapes labelled and in my tapelist - last night's backup used #14 and asked for #15 for tonight. From what I've read before, amdump will use the last tape in the tapelist marked reuse (incidentally, every tape in my tapelist is marked reuse - is that correct ?) so what exactly does the tapecycle parameter do ? If there are less than tapecycle tape in the tapelist will it force amdump to ask for a new one ? A second (actually, third) before I lick the envelope :-) I've got a large holding disk so I may well change runspercycle to 14 and run amdump every day and then run amflush on Monday in order to have as many backups as possible (and people may sometimes do some work at weekends). If I do that and the amount of data in the two holding directories on Monday is small enough, is it OK to amflush ALL to one tape or am I better off flushing each day to a separate tape, which would give the same nett result as if the tapes had been put in the drive over the weekend ? Regards, Niall O Broin
