amcheck, client problem: [could not access... : Permission denied]
I had to split a big partition of a Linux fileserver into several part due to the limit of Amanda backing up filesystem bigger than the tape size. I simply have done it specifying some subdirectories in the disklist file; amdump is working well, excluding the following amcheck error: - amcheck result: Amanda Tape Server Host Check - Server check took 156.049 seconds Amanda Backup Client Hosts Check ERROR: hal: [could not access /home3/ale/Storage/segmentoC (/home3/ale/Storage/segmentoC): Permission denied] ERROR: hal: [could not access /home3/ale/Storage/segmentoB (/home3/ale/Storage/segmentoB): Permission denied] ERROR: hal: [could not access /home3/ale/Storage/segmentoA (/home3/ale/Storage/segmentoA): Permission denied] ERROR: hal: [could not access /home3/ale/Privato (/home3/ale/Privato): Permission denied] ERROR: hal: [could not access /home3/ale/Lavoro (/home3/ale/Lavoro): Permission denied] . (brought to you by Amanda 2.4.2p2) - amcheck result Thank you in advance for any reply. Best regards, Alessandro Prete ICT manager Polytechnic of Turin, branch of Alessandria / CdCMP phone: +39 (0)131-22.93.40 eMail: [EMAIL PROTECTED] CdCMP website: http://www.polial.polito.it/cdc --- For every complex problem there is a simple solution. And it is wrong. ---
sorry for receipt confirmation request
Samba fail - Irix time out
I'm running an Amanda server on a Linux (RH 7.2) machine to backup about twenty client. Every time a Win client fail *for some wrong share permission* (smbclient can enter but not browse the network resource) and not for a time out, a time-out of an Irix 6.5.15 client, waiting for an ack from the server, occurs. As soon as samba works, Irix client starts over working well. I don't understand the relation that is created among the two clients and the server, keeping in mind that if a Unix client doesn't work for some reason, Irix correctly works as it should. Rough Amanda topology: Clients and server work on the same subnet. Server: Linux RH 7.0, Amanda 2.4.2p2 Clients causing problems: Windows hosts sharing directories with wrong permissions Client having problem: Indigo 2 IRIX 6.5.15, Amanda 2.4.2p2 Do you have some suggestions ? Thank you in advance for any reply. Best regards, Alessandro Prete ICT manager Polytechnic of Turin, branch of Alessandria / CdCMP phone: +39 (0)131-22.93.40 eMail: [EMAIL PROTECTED] CdCMP website: http://www.polial.polito.it/cdc
exec command before backup ?
Hi, I would like to execute some commands on a amanda-client before a backup. Can amanda do this ? (i need to do a mysqldump and only backup the dump, not data in use) -- Fabien Seisen [EMAIL PROTECTED] dakol@opn GnuPG ID 1024D/112189D8
Travan 10/20GB ATAPI (IDE) use with Amanda and amlabel
Hi All, I've recently intalled a Travan 20GB tape drive ATAPI (IDE) (M/N: STT32A), and I'm running Linux 7.1 with Amanda 2.4.2p2. I'm yet to get it to work. We are able to see the /dev/nht0 device and can rewind, etc... We cannot amlabel or do amcheck. The software which comes with the tape drive only supports Microsoft Windows and there is no mention of Linux in the documentation or on their website. I plug the tape drive into a Windows machine and install backup-exec and it works, but I had to first media initialize. I am trying to label the tape using amlabel, with the tape device in the config file set to nht0 instead of ht0, ie non-rewinding, but the message received is: rewinding, reading label, no label found rewinding, writing label archive1, checking label amlabel: no label found, are you sure /dev/nht0 is non-rewinding? when I use amcheck, the message is: ERROR:/dev/nht0:not an amanda tape (expecting a new tape) What am I not seeing/checking? Where does Linux get the driver for the tape drive from if I haven't installed the software that comes with it. Does the Red Hat package come with a driver? The autodetect on startup detects a Seagate STT2A, but on the drive it says STT32A. Am I way off ? Please Help! System Instability - The resultant problem since I've installed the tape drive is our Linux file server is freezing up, forced rebooting results which is obviously a no-no. On shutdown, I see a message: Sending all processes the TERM signal... Sending all processes the KILL signal... md: recovery thread got woken up.. md: recovery thread finished... mdrecoveryd(8) flushing signals. ?Is this of importance to me? The system often freezes at this point. The other message I receive comes up at random intivals, sometimes when the tape's activity light is on with no activity is happening, and often causes the computer to freeze: ide-tape: hdd: cleanup_module( ) called while still busy. Regards Trevor Fraser [EMAIL PROTECTED]
Re: exec command before backup ?
you could change the cron entry on the amanda server to remotely execute the mysqldumps with ssh. #!/bin/sh LIST_OF_HOSTS=foo bar baz for host in $LIST_OF_HOSTS; do ssh $host /usr/local/sbin/do-mysql-dump.sh done /usr/sbin/amdump DailySet1 -ben On Mon, 10 Jun 2002, Fabien SEISEN wrote: Hi, I would like to execute some commands on a amanda-client before a backup. Can amanda do this ? (i need to do a mysqldump and only backup the dump, not data in use) -- Fabien Seisen [EMAIL PROTECTED] dakol@opn GnuPG ID 1024D/112189D8
Re: amcheck, client problem: [could not access... : Permissiondenied]
it sounds like amcheck is getting run with the wrong permissions.. can you check the permissions, and paste it for both amcheck and amdump? -ben On Mon, 10 Jun 2002, Alessandro Prete wrote: I had to split a big partition of a Linux fileserver into several part due to the limit of Amanda backing up filesystem bigger than the tape size. I simply have done it specifying some subdirectories in the disklist file; amdump is working well, excluding the following amcheck error: - amcheck result: Amanda Tape Server Host Check - Server check took 156.049 seconds Amanda Backup Client Hosts Check ERROR: hal: [could not access /home3/ale/Storage/segmentoC (/home3/ale/Storage/segmentoC): Permission denied] ERROR: hal: [could not access /home3/ale/Storage/segmentoB (/home3/ale/Storage/segmentoB): Permission denied] ERROR: hal: [could not access /home3/ale/Storage/segmentoA (/home3/ale/Storage/segmentoA): Permission denied] ERROR: hal: [could not access /home3/ale/Privato (/home3/ale/Privato): Permission denied] ERROR: hal: [could not access /home3/ale/Lavoro (/home3/ale/Lavoro): Permission denied] . (brought to you by Amanda 2.4.2p2) - amcheck result Thank you in advance for any reply. Best regards, Alessandro Prete ICT manager Polytechnic of Turin, branch of Alessandria / CdCMP phone: +39 (0)131-22.93.40 eMail: [EMAIL PROTECTED] CdCMP website: http://www.polial.polito.it/cdc --- For every complex problem there is a simple solution. And it is wrong. ---
Hmmmm...all of a sudden I'm strange
Which wouldn't be so shocking, if it were just me. But it's my amanda backup report, and that makes me uncomfortable. Is this something to worry about? I can see the index file in question, and it looks okay. What's the deal with the .tmp? FAILED AND STRANGE DUMP DETAILS: /-- slaw /home lev 0 STRANGE sendbackup: start [slaw:/home level 0] sendbackup: info BACKUP=/bin/gtar sendbackup: info RECOVER_CMD=/bin/gtar -f... - sendbackup: info end ? gtar: ./amanda/uadaily/index/in-db2.unterlaw.com/_var_www/20020608_0.gz.tmp: Warning: Cannot stat: No such file or directory | Total bytes written: 15011840 (14MB, 3.6MB/s) sendbackup: size 14660 sendbackup: end \ Rebecca A. Pakish Systems Administrator Unterberg Associates, P.C. (219) 736-5579 ext. 184
Re: Travan 10/20GB ATAPI (IDE) use with Amanda and amlabel
I'm not sure why the travan tape is not working properly.. you may want to upgrade to Redhat 7.3 for both driver issues, and security reasons.. can I say that again.. UPGRADE :) as far as your system instability, the mdrecovery process is a program that monitors raid disks. it's only needed (if i remember right) when you have a raid disk fail, and it's required to do re-mirroring, or parity rebuilds. as to why it locks up at that point, is a difficult question.. it may just be stalled.. how long have you let it go before powering down the machine? it sounds to me like you have something odd going in the kernel.. if I remember right, 7.1 comes with either a 2.2, or a really old 2.4 kernel.. I suggest getting a copy of 7.3 from someplace like www.cheapbytes.com for $5, and upgrading the system. -ben On Mon, 10 Jun 2002, Trevor Fraser wrote: Hi All, I've recently intalled a Travan 20GB tape drive ATAPI (IDE) (M/N: STT32A), and I'm running Linux 7.1 with Amanda 2.4.2p2. I'm yet to get it to work. We are able to see the /dev/nht0 device and can rewind, etc... We cannot amlabel or do amcheck. The software which comes with the tape drive only supports Microsoft Windows and there is no mention of Linux in the documentation or on their website. I plug the tape drive into a Windows machine and install backup-exec and it works, but I had to first media initialize. I am trying to label the tape using amlabel, with the tape device in the config file set to nht0 instead of ht0, ie non-rewinding, but the message received is: rewinding, reading label, no label found rewinding, writing label archive1, checking label amlabel: no label found, are you sure /dev/nht0 is non-rewinding? when I use amcheck, the message is: ERROR:/dev/nht0:not an amanda tape (expecting a new tape) What am I not seeing/checking? Where does Linux get the driver for the tape drive from if I haven't installed the software that comes with it. Does the Red Hat package come with a driver? The autodetect on startup detects a Seagate STT2A, but on the drive it says STT32A. Am I way off ? Please Help! System Instability - The resultant problem since I've installed the tape drive is our Linux file server is freezing up, forced rebooting results which is obviously a no-no. On shutdown, I see a message: Sending all processes the TERM signal... Sending all processes the KILL signal... md: recovery thread got woken up.. md: recovery thread finished... mdrecoveryd(8) flushing signals. ?Is this of importance to me? The system often freezes at this point. The other message I receive comes up at random intivals, sometimes when the tape's activity light is on with no activity is happening, and often causes the computer to freeze: ide-tape: hdd: cleanup_module( ) called while still busy. Regards Trevor Fraser [EMAIL PROTECTED]
Re: cruft file ?
Stephen, On Fri, 7 Jun 2002, Stephen Carville wrote: |On Fri, 7 Jun 2002, Marcelo Souza wrote: | |- What those ignoring cruft file mean? | |The directory you have designated as holdingdisk probaly has some |files amanda does not recognize in it. But what it tell as cruft file are part of the backup. This volume is backed up to the holding disk, and later is flushed to tape. Amanda is spliting the volume in some files that amflush call cruft file. Can it mean that am I loosing data that is not flushed to tape? |- NOTES: |- amflush: host1.sd2a.0.1: ignoring cruft file. |- amflush: host1.sd2a.0.2: ignoring cruft file. |- amflush: host1.sd2a.0.3: ignoring cruft file. |- amflush: host1.sd2a.0.4: ignoring cruft file. |- amflush: host1.sd2a.0.5: ignoring cruft file. |- taper: tape host1_MAIL3 kb 11181824 fm 1 [OK] |- |- |- DUMP SUMMARY: |- DUMPER STATS TAPER STATS |- HOSTNAME DISK L ORIG-KB OUT-KB COMP% MMM:SS KB/s MMM:SS KB/s |- -- -- -- |- host1 sd2a 0 11181824 11181824 -- N/AN/A 141:40 1315.5 |- |- (brought to you by Amanda version 2.4.1p1) - Marcelo
Re: Hmmmm...all of a sudden I'm strange
On Mon, 10 Jun 2002 at 10:15am, Rebecca Pakish wrote ? gtar: ./amanda/uadaily/index/in-db2.unterlaw.com/_var_www/20020608_0.gz.tmp: Warning: Cannot stat: No such file or directory This file disappeared in the middle of the gtar run. No big deal. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
tape changer
Hello all I have a Overlan Ait-3 Library Pro. I beileve I have everything configured. When I attempt to issue a command such as amtape config update I always get this error could not load slot 2: Drive not ready after 120 seconds, rewind said /dev/st0: Input/output error I also get the same error when I run amlabel I am using the chg-zd-mtx Tape changer with one config If anyone can provide any assitance I'd greeatly appriciate it. Craig Hancock
Re: Hmmmm...all of a sudden I'm strange
Its nothing to worry about. It just means that file was in the process of being compressed when tar got the directory contents but was finished (and the tmp file gone) when tar actually tried to add it to the archive. You should have the uncompressed version of the file on the tape and will get the compressed version backed up on the next run. Frank --On Monday, June 10, 2002 10:15:04 -0500 Rebecca Pakish [EMAIL PROTECTED] wrote: Which wouldn't be so shocking, if it were just me. But it's my amanda backup report, and that makes me uncomfortable. Is this something to worry about? I can see the index file in question, and it looks okay. What's the deal with the .tmp? FAILED AND STRANGE DUMP DETAILS: /-- slaw /home lev 0 STRANGE sendbackup: start [slaw:/home level 0] sendbackup: info BACKUP=/bin/gtar sendbackup: info RECOVER_CMD=/bin/gtar -f... - sendbackup: info end ? gtar: ./amanda/uadaily/index/in-db2.unterlaw.com/_var_www/20020608_0.gz.tmp: Warning: Cannot stat: No such file or directory | Total bytes written: 15011840 (14MB, 3.6MB/s) sendbackup: size 14660 sendbackup: end \ Rebecca A. Pakish Systems Administrator Unterberg Associates, P.C. (219) 736-5579 ext. 184 -- Frank Smith[EMAIL PROTECTED] Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
tape changer chg-scsi problem
Hi, I've being using amanda for a couple of years. Recently I upgrade amanda from 2.4.1 to 2.4.2p2. I also upgrade the clients to 2.4.2p2. Now I have problems with chg-scsi. I red once that the new version of amanda's chg-scsi changer supports Exabyte's X220 library. I configured amanda.conf with chg-scsi: tpchanger /usr/lib/amanda/chg-scsi tapedev 0 rawtapedev /dev/null # the raw device to be used (ftape only) changerfile /etc/amanda/Daily/chg-scsi.conf changerdev /dev/sg0 In chg-scsi.conf I set eject 0 (to use eject), changer to sg0, tape to nst0, cleancart to 1, rotate tapes between 10 and 19. Here is my chg-scsi.conf: number_configs 1 eject 2 # Tapedrives need an eject command sleep 25 # Seconds to wait until the tape gets ready changerdev /dev/sg0 # Next comes the data for drive 0 config 0 # Daily drivenum0 dev /dev/nst0 # the device that is used for the tapedrive 0 scsitapedev /dev/sg0 tapestatus /home/amanda/Daily/tapestatus labelfile /home/amanda/Daily/labelfile startuse10 enduse 19 statfile/home/amanda/Daily/statfile cleancart 1 cleanfile /home/amanda/Daily/tape-clean usagecount /home/amanda/Daily/totaltime 'chg-scsi -status' seems to recognize that the library has 21 elements: DecodeModeSense : Element Address Assignment Page Medium Transport Element Address 86 Number of Medium Transport Elements 1 First Storage Element Address 0 Number of Storage Elements 21 First Import/Export Element Address 0 Number of ImportExport Elements0 First Data Transfer Element Address 82 Number of Data Transfer Elements 1 but 'amtape config update' doesn't unload a tape to the slot from where it was previously loaded but unload it always to slot 1 (that is declared as cleancart and shouldn't be used). Than I get 'Input/output error': [amanda@craft83 amanda]$ amtape Daily update amtape: scanning all 10 slots in tape-changer rack: slot 0: date 20020520 label Daily_0039 slot 1: rewinding tape: Input/output error What is wrong with my configuration? Thanks in advance, Ted P.S. Here is the output from 'chg-scsi -status all' [amanda@craft83 amanda]$ chg-scsi -status all Ident = C1553A, type = HP Auto Loader [C1553A] Ident = EXB-10e, type = Exabyte Robot [EXB-10e] Ident = EXB-120, type = Exabyte Robot [EXB-120] Ident = EXB-230D, type = Exabyte Robot [EXB-230D] Ident = EXB-85058HE-, type = Exabyte Tape [EXB-85058HE-] Ident = TDS 1420, type = Tandberg Robot (TDS 1420) Ident = DLT7000, type = DLT Tape [DLT7000] Ident = DLT4000, type = DLT Tape [DLT4000] Ident = VLS DLT, type = ADIC VLS DLT Library [VLS DLT] Ident = VLS SDX, type = ADIC VLS DLT Library [VLS SDX] Ident = Scalar DLT 448, type = ADIC DLT 448 [Scalar DLT 448] Ident = 215, type = Spectra Logic TreeFrog[215] Ident = generic, type = Generic driver tape/robot [generic] Address Type Status From --- 086 MTE Empty -001 000 STE Empty 0082 001 STE Full -001 002 STE Full -001 003 STE Full -001 004 STE Full -001 005 STE Full -001 006 STE Full -001 007 STE Full -001 008 STE Full -001 009 STE Full -001 010 STE Full 0082 011 STE Full -001 012 STE Full -001 013 STE Full -001 014 STE Full -001 015 STE Full -001 016 STE Full 0082 017 STE Full 0082 018 STE Empty 019 STE Full -001 020 STE Full -001 082 DTE Full 0018 Sense Status from robot: # START DecodeExtSense Extended Sense # START DecodeSense Sense Keys ErrorCode 70 Valid 0 ASC 00 ASCQ 00 Sense key 00 No Sense Log Parameter Page Code 00 Log Parameter Code 00 Underrun/Overrun Counter00 Read/Write Error Counter1 Remaing 1024 byte tape blocks -128 Tracking Retry Counter 00 Read/Write Retry Counter12 Fault Sympton Code 00 Sense Status from tape: # START DecodeExtSense Extended Sense # START DecodeSense Sense Keys ErrorCode 70 Valid 0 ASC 00 ASCQ 00 Sense key 00 No Sense Log Parameter Page Code 00 Log Parameter Code 00 Underrun/Overrun Counter00 Read/Write Error Counter66 Formatted Buffer parity Error Error Counter Overflow TapeMotion Error Logical Beginning of tape
Re: 2 tape strategy
you also need to update the cron entries so that it only runs on the days you want it to.. read the crontab man page, and run the apropriate crontab command for your system. (i can't tell you any more, because you didn't say what OS and version you're running) -ben On Mon, 10 Jun 2002, Fernan Aguero wrote: Hi! I am running amanda-2.4.2p2 succesfully with 2 hosts (1 server, 1 client). All disklist entries (level 0 dump) fit well on a DDS-4 40GB using software compression (about 24 GB now). I have only 4 tapes in use and momentarily just 2 (I am keeping 2 of them as archival tapes as I am burning some of the data on CDs, and will return them to the tapecycle as soon as they get done). This is how I've set my conf file: dumpcycle 7 runspercycle 2 tapecycle 2 tapes Based on this I'd expected amanda to run just twice every 7 days. However, amanda still runs everyday. Is it possible to do this? (if i set a dumpcycle of 30 and runspercycle of 1, would amanda run just once per month?) I am currently forcing level 0 dumps in every run because otherwise amanda would almost always overwrite the only level 0 dump for a filesystem. Since all lev 0 dumps are only a few megabytes ( 100 MB), is it possible to have amanda append these to the end of each tape, thus leaving the level 0 dumps untouched? Currently I have this problem because on weekends amanda would run without being forced to do a lev 0 dump and thus would leave things in the holding disk that need to be flushed on Monday. That day, I will be left with only 1 level 0 dump in tape for each filesystem. Thanks in advance, Fernan PS: I know that just getting more tapes would be the ideal solution, but that is not economically possible in the short term. I'll appreciate comments on how to make amanda run as tapewise as it can, given my shortage of tape media. -- | F e r n a n A g u e r o | B i o i n f o r m a t i c s | | [EMAIL PROTECTED] | genoma.unsam.edu.ar |
Re: 2 tape strategy
--On Monday, June 10, 2002 17:12:08 -0300 Fernan Aguero [EMAIL PROTECTED] wrote: Hi! I am running amanda-2.4.2p2 succesfully with 2 hosts (1 server, 1 client). All disklist entries (level 0 dump) fit well on a DDS-4 40GB using software compression (about 24 GB now). I have only 4 tapes in use and momentarily just 2 (I am keeping 2 of them as archival tapes as I am burning some of the data on CDs, and will return them to the tapecycle as soon as they get done). This is how I've set my conf file: dumpcycle 7 runspercycle 2 tapecycle 2 tapes Based on this I'd expected amanda to run just twice every 7 days. However, amanda still runs everyday. Is it possible to do this? (if i set a dumpcycle of 30 and runspercycle of 1, would amanda run just once per month?) Amanda runs whenever you (or cron) make it run. 'runspercycle' is just for planner's benefit so it can schedule appropriate dump levels. IF you only want Amanda to just run twice a week, just have it scheduled that way in cron. I am currently forcing level 0 dumps in every run because otherwise amanda would almost always overwrite the only level 0 dump for a filesystem. Since all lev 0 dumps are only a few megabytes ( 100 MB), is it possible to have amanda append these to the end of each tape, thus leaving the level 0 dumps untouched? Currently I have this problem because on weekends amanda would run without being forced to do a lev 0 dump and thus would leave things in the holding disk that need to be flushed on Monday. That day, I will be left with only 1 level 0 dump in tape for each filesystem. If you want more level 1s and have enough holding disk, you could run Amanda daily without a tape (and adjust your config file accordingly), and amflush twice a week. There is the added risk of losing your backups if your holding disk fails. Frank Thanks in advance, Fernan PS: I know that just getting more tapes would be the ideal solution, but that is not economically possible in the short term. I'll appreciate comments on how to make amanda run as tapewise as it can, given my shortage of tape media. -- | F e r n a n A g u e r o | B i o i n f o r m a t i c s | | [EMAIL PROTECTED] | genoma.unsam.edu.ar | -- Frank Smith[EMAIL PROTECTED] Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
Re: 2 tape strategy
+[ Ben Kochie ([EMAIL PROTECTED]) dijo sobre Re: 2 tape strategy: | | you also need to update the cron entries so that it only runs on the days | you want it to.. read the crontab man page, and run the apropriate crontab | command for your system. (i can't tell you any more, because you didn't | say what OS and version you're running) | | -ben Yes! Thanks Ben ... I completely forgot about the cron job. However, now my question is: how does the interaction of cron job and dumpcycle/runspercycle work? What is the meaning of runspercycle in amanda.conf if the actual 'runs per cycle' are set through the cron job? i.e. if i set a dumpcycle of 7 days (meaning I want at least 1 level 0 dump of each fs per week) and runspercycle to 1, I would expect to have amanda run only once per week and do a full dump of every fs. However if my cron job is set to run amcheck + amdump every day, I suppose amanda will run daily (as it's happening now) and will try to dump daily to tape. (Why doesn't amanda skip some days, even if called from a cron job?). Fernan | | On Mon, 10 Jun 2002, Fernan Aguero wrote: | | Hi! | | I am running amanda-2.4.2p2 succesfully with 2 hosts (1 server, 1 | client). All disklist entries (level 0 dump) fit well on a DDS-4 40GB | using software compression (about 24 GB now). | | I have only 4 tapes in use and momentarily just 2 (I am keeping 2 of them as | archival tapes as I am burning some of the data on CDs, and will | return them to the tapecycle as soon as they get done). | | This is how I've set my conf file: | | dumpcycle 7 | runspercycle 2 | tapecycle 2 tapes | | Based on this I'd expected amanda to run just twice every 7 days. | However, amanda still runs everyday. | Is it possible to do this? (if i set a dumpcycle of 30 and | runspercycle of 1, would amanda run just once per month?) | | I am currently forcing level 0 dumps in every run because otherwise | amanda would almost always overwrite the only level 0 dump for a | filesystem. | | Since all lev 0 dumps are only a few megabytes ( 100 MB), is it | possible to have amanda append these to the end of each tape, thus | leaving the level 0 dumps untouched? | | Currently I have this problem because on weekends amanda would run | without being forced to do a lev 0 dump and thus would leave things in | the holding disk that need to be flushed on Monday. That day, I will | be left with only 1 level 0 dump in tape for each filesystem. | | Thanks in advance, | | Fernan | | PS: I know that just getting more tapes would be the ideal solution, | but that is not economically possible in the short term. I'll | appreciate comments on how to make amanda run as tapewise as it can, | given my shortage of tape media. | | -- | | | F e r n a n A g u e r o | B i o i n f o r m a t i c s | | | [EMAIL PROTECTED] | genoma.unsam.edu.ar | | | | | +] -- | F e r n a n A g u e r o | B i o i n f o r m a t i c s | | [EMAIL PROTECTED] | genoma.unsam.edu.ar |
Re: amrestore program
On Tue, Jun 04, 2002 at 09:44:12PM -0600, Trevor Morrison wrote: Hi, When I do my backups I use the GNU tar for my root-tar dumptype. While I was trying to restore a file tonight I would get the following error: tar: ./trevor/trevor: Cannot open: Not a directory tar: Error exit delayed form previous errors extract_list - child returned non-zero status: 2 trevor is just regular file in my /home/trevor directory. I first changed into that directory before starting amrecover. What am I doing wrong? TIA. tar try to create the 'trevor' directory in /home/trevor but you already have a trevor file there. You should cd to the mount point before you use amrecover. Jean-Louis PS. That's always a bad idea to have the same name for two different things. -- 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: 2 tape strategy
check the other replies to your message.. some more detailed info is included. the dumpcycle/runspercycle is there to help the planner along.. nothing more.. remember, the amanda planner is psudo-smart.. it tries to predict which disk should get their full backups which days. the cron job is there to just process a current dump.. allowing you to decide exactly when dumps should occur, not allowing the planer to try and figure it out for you. -ben On Mon, 10 Jun 2002, Fernan Aguero wrote: +[ Ben Kochie ([EMAIL PROTECTED]) dijo sobre Re: 2 tape strategy: | | you also need to update the cron entries so that it only runs on the days | you want it to.. read the crontab man page, and run the apropriate crontab | command for your system. (i can't tell you any more, because you didn't | say what OS and version you're running) | | -ben Yes! Thanks Ben ... I completely forgot about the cron job. However, now my question is: how does the interaction of cron job and dumpcycle/runspercycle work? What is the meaning of runspercycle in amanda.conf if the actual 'runs per cycle' are set through the cron job? i.e. if i set a dumpcycle of 7 days (meaning I want at least 1 level 0 dump of each fs per week) and runspercycle to 1, I would expect to have amanda run only once per week and do a full dump of every fs. However if my cron job is set to run amcheck + amdump every day, I suppose amanda will run daily (as it's happening now) and will try to dump daily to tape. (Why doesn't amanda skip some days, even if called from a cron job?). Fernan | | On Mon, 10 Jun 2002, Fernan Aguero wrote: | | Hi! | | I am running amanda-2.4.2p2 succesfully with 2 hosts (1 server, 1 | client). All disklist entries (level 0 dump) fit well on a DDS-4 40GB | using software compression (about 24 GB now). | | I have only 4 tapes in use and momentarily just 2 (I am keeping 2 of them as | archival tapes as I am burning some of the data on CDs, and will | return them to the tapecycle as soon as they get done). | | This is how I've set my conf file: | | dumpcycle 7 | runspercycle 2 | tapecycle 2 tapes | | Based on this I'd expected amanda to run just twice every 7 days. | However, amanda still runs everyday. | Is it possible to do this? (if i set a dumpcycle of 30 and | runspercycle of 1, would amanda run just once per month?) | | I am currently forcing level 0 dumps in every run because otherwise | amanda would almost always overwrite the only level 0 dump for a | filesystem. | | Since all lev 0 dumps are only a few megabytes ( 100 MB), is it | possible to have amanda append these to the end of each tape, thus | leaving the level 0 dumps untouched? | | Currently I have this problem because on weekends amanda would run | without being forced to do a lev 0 dump and thus would leave things in | the holding disk that need to be flushed on Monday. That day, I will | be left with only 1 level 0 dump in tape for each filesystem. | | Thanks in advance, | | Fernan | | PS: I know that just getting more tapes would be the ideal solution, | but that is not economically possible in the short term. I'll | appreciate comments on how to make amanda run as tapewise as it can, | given my shortage of tape media. | | -- | | | F e r n a n A g u e r o | B i o i n f o r m a t i c s | | | [EMAIL PROTECTED] | genoma.unsam.edu.ar | | | | | +] -- | F e r n a n A g u e r o | B i o i n f o r m a t i c s | | [EMAIL PROTECTED] | genoma.unsam.edu.ar |
kbps configuration directive
I'm curious about the `kbps' keyword used in Amanda configuration files, especially for determining network bandwidth.. Is that measured in kilobytes/second or kilobits/second? Also, does specifying an amount for `netusage' actually affect anything -- does Amanda rate-limit data coming in? Just curious (and a little confused), thanks.. -- Mike Hicks [mailto:[EMAIL PROTECTED]] Unix Support Assistant| Carlson School of Management Office: 1-160 Phone: 6-7909 | University of Minnesota msg12733/pgp0.pgp Description: PGP signature
New User, config issues
Hello, I just installed amanda without issue and have been reading the man pages, faq, and egroups archive. I still have some issues. First some details on my setup. I have a Linux server with an internal tape drive. I have 8 tapes (DLT 20/40) and want at least one full backup per week. I read that amanda always writes from the start of the tape (no appending). That is fine, it's what I do now with tar and cron. I'm not here to change tapes on the weekend so backups will run only on the weekdays (which I will schedule in cron). Right now, I will be backing up only the local machine, no clients. Also, I don't have much drive space to buffer the backups to disk before they go to tape (27GB of drive space, about 25 full). Now, what I don't really understand is some of the config file options. I want at least one full backup each week. My current config is: dumpcycle 7 days runspercycle 7 tapecycle 8 tapes But I'm not sure if that should be dumpcycle 5 days runspercycle 5 tapecycle 8 tapes since I will only be doing backups on the weekends. With these configs, I'm not sure when amanda will do full backups. This is a bit confusing to me. Next, I have the following tape device: Compaq 20/40-GB DLT Drive http://www.compaq.com/products/storageworks/dlt2040/index.html http://www.compaq.com/products/quickspecs/10234_div/10234_div.html I'm not sure which tapetype to use (or how to configure my own) from the specs above, there doesn't seem to be a match in the files and the specs don't give me the needed values. Is there a default DLT setting? Thanks for any help you can provide. Mike
Backup to disk drive (tapeless operation)
I read somewhere that the ability to back up directly to a hard drive rather than tape exists in 2.4.3. Could somebody point me in the right direction on how to set this up? I looked through the amanda.conf and searched through the documentation, but couldn't find anything that would relate to this. Thank you. Sincerely, Jeremy Koski
Re: Travan 10/20GB ATAPI (IDE) use with Amanda and amlabel
Hello Trevor
I wish I had the answer for you but unfortunately after getting amanda to
work it has now reverted back to what you are experiencing. And worse still
I'm not sure why :( All I know is that I had it working for a few days, I
then rebooting the server and haven't had it working since.
Below are some snippets from my amanda.conf file. Please let me know if you
have any joy with getting this resolved.
runtapes 1 # number of tapes to be used in a single run of
amdump
tapedev /dev/nht0 # the no-rewind tape device to be used
tapetype STT2A # what kind of tape it is (see tapetypes
below)
labelstr ^Tape[0-9][0-9]*$# label constraint regex: all tapes must
match
define tapetype STT2A {
comment just produced by tapetype program
length 9500 mbytes
filemark 103 kbytes
speed 914 kbytes
}
Although the message from another list member states;
As far as i recall the safest way to use this ide-tapes under linux
is activating scsi-emulation for the drive and then acess it via
/dev/nst[0..n]
the linux ide-tape-driver seems to have some problems with the
way amanda uses tape drives. at least with some tape drives.
I have not tried this, which could be the answer.
Regards
Brad
- Original Message -
From: Trevor Fraser [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Monday, June 10, 2002 6:02 PM
Subject: Travan 10/20GB ATAPI (IDE) use with Amanda and amlabel
Hi All,
I've recently intalled a Travan 20GB tape drive ATAPI (IDE) (M/N:
STT32A), and I'm running Linux 7.1 with Amanda 2.4.2p2. I'm yet to get
it to work. We are able to see the /dev/nht0 device and can rewind, etc...
We cannot amlabel or do amcheck.
The software which comes with the tape drive only supports Microsoft
Windows
and there is no mention of Linux in the documentation or on their website.
I
plug the tape drive into a Windows machine and install backup-exec and it
works, but I had to first media initialize.
I am trying to label the tape using amlabel, with the tape device in the
config file set to nht0 instead of ht0, ie non-rewinding, but the message
received is:
rewinding, reading label, no label found
rewinding, writing label archive1, checking label
amlabel: no label found, are you sure /dev/nht0 is non-rewinding?
when I use amcheck, the message is:
ERROR:/dev/nht0:not an amanda tape
(expecting a new tape)
What am I not seeing/checking?
Where does Linux get the driver for the tape drive from if I haven't
installed the software that comes with it. Does the Red Hat package come
with a driver? The autodetect on startup detects a Seagate STT2A, but
on
the drive it says STT32A. Am I way off ? Please Help!
System Instability
-
The resultant problem since I've installed the tape drive is our Linux
file
server is freezing up, forced rebooting results which is obviously a
no-no.
On shutdown, I see a message:
Sending all processes the TERM signal...
Sending all processes the KILL signal... md: recovery thread got woken
up..
md: recovery thread finished...
mdrecoveryd(8) flushing signals.
?Is this of importance to me? The system often freezes at this point.
The other message I receive comes up at random intivals, sometimes when
the
tape's activity light is on with no activity is happening, and often
causes
the computer to freeze:
ide-tape: hdd: cleanup_module( ) called while still busy.
Regards
Trevor Fraser
[EMAIL PROTECTED]
