Samba issues
I found out that if I run more than one samba dump (smbclient) at once, the smb jobs get stucked. If I run them serialy (spindle) it works much better (probaby ok). There must be some issues with multiple simultaneus smb jobs, but I cant tell what is wrong. Also the jobs have to be a little bit larger (few GB at least). regards, gregor
Re: Samba issues
There must be some issues with multiple simultaneus smb jobs, but I cant tell what is wrong. Not using amanda, but Samba in general, I found that when I ran a domain controller that opened shares from one of its clients (a Win98 box) things would crawl until I killed samba PDC or smbmount.. Definitely sounds like the SMB software wrestling with itself. Maybe someone will have a solution to both situations..
Muhteþem Hazýr Web Tasarýmlarý (Template's)
Webmaster'larn dikkatine, Hazr web sitesi tasarmlar, web sitenizi hazrlamak amacyla retilmi olup internete kaliteli ve profesyonel bir giri yapmak iin sizi beklemektedir... * Orjinal Photoshop (.psd) veya (.png) dosyas * Tasarmda kullanlan fontlar * Varsa extra Jpeg resimler * HTML formatnda Ana Sayfa * HTML formatnda Sayfa * Roll Over butonlar * Orjinal Profesyonel Tasarmlar Tm tasarmlar Macromedia Dreamweaver, Microsoft Frontpage, Adobe Golive ve dier HTML editrleri ile uyumludur... rnlerimize ilikin olarak web sitemizi ziyaret edebilirsiniz. (Ksa bir sre iin fiyatlar indirdik, 1 hazr tasarm 20 Milyon TL) Sayglarmzla, Disket.NET Web Design Studio www.disket.net
Re: /etc/exclude.gtar is ignored!
Turgut Kalfaoglu wrote:
Hi there. I backup a system, and I notice that it complains about 'unusual
or strange backup result' during the email. These are files I forgot to
exclude; things like logs, that change during the backup.
So, I added them to /etc/exclude.gtar which is defined properly in
amanda.conf in 'server-tar', the exclude.gtar is stored at the client, and
I am running the client as root (so that it has access to everything), and
amcheck shows no problems. yet, every day I keep getting complaints about
the same files; like ./apache/cgi-bin/openwebmail/etc/sessions/blah has
changed. well, in exclude.gtar I have:
./apache/cgi-bin/openwebmail/etc/sessions
and I assumed that this should have stopped everything under 'sessions'
from being backed up?
I checked the wording, in fact, I did cut/paste from the error emails.
I'm obviously missing something! Can anyone help?
Obviously there is something wrong. But besides telling us everything
is allright, you didn't prove it to us.
Show the exact errors, show the exact configs, instead of saying they
are defined properly. Also mentioning the program version(s) can help
sometimes.
In the worst case, you get a second opinion about the correctness,
and in the best case, we notice the error.
One strange thing already is that you run the client as root. That's
not the way it should be done. In fact there is code in amandad.c (line
148-167) to avoid this. The runtar and rundump programs in libexec
should be installed suid-root however, just to overcome the limitation
about access to all files.
But I guess this has nothing to do with the exclude problem.
A common error with excludes is the omission of the list keyword
when specifying a file with a list of excludes.
hour.host /the/directory {
comp-user-tar
exclude list /etc/exclude.gtar
}
Without the list keyword, the file itself would be excluded, which
actually never matches, because the patterns need to start with a dot.
--
Paul @ Home
Re: Problem configuring amanda before compiling
On Tuesday 11 November 2003 00:11, Dana Bourgeois wrote: Thanks. Doing a 'make distclean' and './configure' didn't change anything but still, thanks. If you didn't give configure a list of options, amanda will be pretty crippled. Because I always have the latest snapshot in service, and I need to have a consistent set of options so that the next snapshot wil run exactly like the last, I long ago cobbled up a script to do the configuration and make. Here it is. --- #!/bin/sh # since I'm always forgetting to su amanda... if [ `whoami` != 'amanda' ]; then echo echo Warning echo Amanda needs to be configured and built by the user amanda, echo but must be installed by user root. echo exit 1 fi make clean rm -f config.status config.cache ./configure --with-user=amanda \ --with-group=disk \ --with-owner=amanda \ --with-tape-device=/dev/nst0 \ --with-changer-device=/dev/sg1 \ --with-gnu-ld --prefix=/usr/local \ --with-debugging=/tmp/amanda-dbg/ \ --with-tape-server=coyote.coyote.den \ --with-amandahosts \ --with-configdir=/usr/local/etc/amanda make --- Adjust what needs to be adjusted of course, and remove the references to the changer if you don't have one. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jon LaBadie Sent: Monday, November 10, 2003 7:25 PM To: [EMAIL PROTECTED] Subject: Re: Problem configuring amanda before compiling On Mon, Nov 10, 2003 at 06:44:27PM -0800, Dana Bourgeois wrote: - I didn't see anything that would remove/invalidate/unconfigure the previous ./configure step). I use make distclean. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax) -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: Problem configuring amanda before compiling
On Tuesday 11 November 2003 00:36, Dana Bourgeois wrote: Answered my own question once I had time to dig a little. You need to have the readline-devel rpm installed as well as the readline rpm. No one came forward about the QIC tape warning but I will know after I configure the disk 'tapes' and run an amdump if it's a problem. If you don't actually have a QIC drive, they can be ignored. Its an option to turn on when compiling a new kernel IIRC. The problem finder isn't amdump per sei, but amcheck, which will scan the system and check the install, reporting warnings (which can be ignoored, and errors that will stop the amdump. Errors must be fixed. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dana Bourgeois Sent: Monday, November 10, 2003 6:44 PM To: [EMAIL PROTECTED] Subject: Problem configuring amanda before compiling On RedHat, I didn't run into this. Now I'm on SsSE8.1 and the ./configure gives me a warning that I don't have a terminal library for interactive amrecover nor do I have the vtblc stuff needed for QIC tape support. Possibly I don't care. I'm going to use the file: driver with this installation and the chg-multi tape changer. I have the readline package installed. Anyone know if I care about these warnings and if so, which packages I need to tickle YaST to install for me? Please believe that I have already tried searching by package name and installed several packages that looked pertinent but didn't make these warnings go away, (ummm... I just reran ./configure - I didn't see anything that would remove/invalidate/unconfigure the previous ./configure step). Clues gratefully accepted. Dana Bourgeois -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: Network recovery with changer
On Mon, Nov 10, 2003 at 03:39:22PM -0800, Stephen Walton wrote: On Mon, 2003-11-10 at 09:44, Paul Bijnens wrote: Stephen Walton wrote: I'm guessing from the responses I'm getting that no one else on the list is even trying to use a changer directly from amrecover? Which version are you using? I was using 2.4.4p1 but after your message I gave the November 7 snapshot a try. I suppose it fixes the problem of the lack of an 'amrecover_check_header' line in my amanda.conf. However, it still doesn't seem to work for me. I'm sure I'm being extraordinarily stupid here, but once more: my amanda.conf has the lines: tpchanger chg-scsi# the tape-changer glue script tapedev 0 changerfile /etc/opt/amanda/daily/chg-scsi-compaq.conf amrecover_changer /dev/nst0 Which of these lines has to change, and what settape command in amrecover do I use, in order to get amrecover to use the changer? The amanda manpage notwithstanding, settape /dev/nst0 with the above setup does not work. Hi Stephen, From amanda man page: amrecover_changer string Default: ''. Amrecover will use the changer if you use 'settape STRING' and that string is the same as the amrecover_changer setting. From amrecover man page: settape [[server]:][tapedev|default] If you want amrecover to use your changer, the tapedev must be equal to the amrecover_changer setting on the server. You can use any string for amrecover_changer, but you must use the same string for the settape command in amrecover. Could you provide your amidxtaped.timestamp.debug file? Jean-Louis -- Jean-Louis Martineau email: [EMAIL PROTECTED] Département IRO, Université de Montréal C.P. 6128, Succ. CENTRE-VILLETel: (514) 343-6111 ext. 3529 Montréal, Canada, H3C 3J7Fax: (514) 343-5834
amanda type type definition
hi, did anybody has a configuration for a HP C5683A C111 DAT Streamer DDS 4 20GB thank´s -- Mit freundlichen Grüßen Michael Hollmann JAWA Management Software GmbH A-8041 Graz, Liebenauer Hauptstraße 200 Tel: ++43 (0)316 403274-13 Fax: ++43 (0)316 403274-10 GSM: ++43 (0)676 4101431 E-Mail: [EMAIL PROTECTED] Web: http://www.jawa.at/ News: http://www.jawa.at/news.php/ Web-Framework für B2B-Applikationen
Amdump Mail Report - Description
Hi, amanda-users, I just talked to a customer about the various times in the Amanda Mail Report. There is the Estimate Time, Run Time, Dump Time, Tape Time I understand them pretty well, but I would like to know if and where the listed entries are described. (Would not hurt to see all the other entries described as well, even if most of them are no-brainers) For example, I only found out the meanings of Dump Time and Run Time via web research (which lead me to some mail of Jon LaBadie asking a similar question ... ;-) ). I looked through some manpages, but could not find it ... Thank you, Stefan G. Weichinger mailto:[EMAIL PROTECTED]
can't backup a disk?
Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen -- === Yu Chen Howard Hughes Medical Institute Chemistry Building, Rm 122 University of Maryland at Baltimore County 1000 Hilltop Circle Baltimore, MD 21250 phone: (410)455-6347 (410)455-2718 fax:(410)455-1174 email: [EMAIL PROTECTED] ===
Re: can't backup a disk?
On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: can't backup a disk?
Thanks, Gene Think that might be the problem. I will try it tonight. So you mean by dump, it should backup everything under /var including /var/spool/mail? But it didn't happen on my system, the dump didn't back it up. That's why I added the entry. Chen On Tue, 11 Nov 2003, Gene Heskett wrote: On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- === Yu Chen Howard Hughes Medical Institute Chemistry Building, Rm 182 University of Maryland at Baltimore County 1000 Hilltop Circle Baltimore, MD 21250 phone: (410)455-6347 (410)455-2718 fax:(410)455-1174 email: [EMAIL PROTECTED] ===
Moving data index data etc. from one config to andother
Following our recent reorg of amanda configs, I've considered moving some data from index, curinfo and possibly the log of one config to the datadirs of another. The object would be to make amrecover think that certain tapes were written using this other config, although they were really dumped by the first one. Has anyone tried this? Will it work at all, and if it does, what exactly do I need to include? - Toralf
Re: can't backup a disk?
On Tuesday 11 November 2003 10:13, Yu Chen wrote: Thanks, Gene Think that might be the problem. I will try it tonight. So you mean by dump, it should backup everything under /var including /var/spool/mail? But it didn't happen on my system, the dump didn't back it up. That's why I added the entry. Chen No, by dump, I meant the program 'dump', as opposed to the program 'tar' (usually gnutar, and it needs to be at least version 1.13-19, plain old 1.13 won't do, its busted, get 1.13-25 from alpha.gnu.org's ftp site) tar should be able to do this, no problem, although if you leave the email agent running full time (I do, I'm on dsl) then its every so often scans of your mailbox at your ISP might occasionally cause a file changed while we read it message from tar if its updateing the mail spool while tar is reading it. However, since you have /var in the disklist, and tar traverses all subdirs below the named one, there must be a more fundamental problem, possibly related to permissions. You did configure and build amanda as the user 'amanda' didn't you? And then become root before installing her? On Tue, 11 Nov 2003, Gene Heskett wrote: On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
includefile directive?
Is anyone here using the includefile directive in their config? How exactly does it work? Does it apply to all config files, or just amanda.conf? What can the file contain - full config info, or just whatever is not set in the file including it? If I have two configs, can I have one amanda.conf include the other and override some of the params, or is it better to let both include a 3rd file (or several different files for the different sections)? - Toralf
Re: can't backup a disk?
--On Tuesday, November 11, 2003 10:35:32 -0500 Gene Heskett [EMAIL PROTECTED] wrote: On Tuesday 11 November 2003 10:13, Yu Chen wrote: Thanks, Gene Think that might be the problem. I will try it tonight. So you mean by dump, it should backup everything under /var including /var/spool/mail? But it didn't happen on my system, the dump didn't back it up. That's why I added the entry. Chen No, by dump, I meant the program 'dump', as opposed to the program 'tar' (usually gnutar, and it needs to be at least version 1.13-19, plain old 1.13 won't do, its busted, get 1.13-25 from alpha.gnu.org's ftp site) tar should be able to do this, no problem, although if you leave the email agent running full time (I do, I'm on dsl) then its every so often scans of your mailbox at your ISP might occasionally cause a file changed while we read it message from tar if its updateing the mail spool while tar is reading it. However, since you have /var in the disklist, and tar traverses all subdirs below the named one, there must be a more fundamental problem, possibly related to permissions. You did configure and build amanda as the user 'amanda' didn't you? And then become root before installing her? On Tue, 11 Nov 2003, Gene Heskett wrote: On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail Notice that /var/spool/mail is a seperate partion than /var. Dump and tar (with the options Amanda calls it with) does not cross filesytems, so a backup of /var on your machine will not backup /var/spool/mail. Frank before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved. -- Frank Smith [EMAIL PROTECTED] Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
backup lasts forever on large fs
I'm trying to backup a home partition with amanda under Sun Solaris 9 with Amanda version 2.4.4p1: Filesystem size used avail capacity Mounted on /dev/md/dsk/d0 102G50G51G50%/home which contains a lot of i-nodes: Filesystem iused ifree %iused Mounted on /dev/md/dsk/d0 2932711 992348123% /home I can write it to tape in ~5-6 hours using plain gtar but amanda seems to be busy for several hours (10+) without writing anything to tape. Is there a workaround to get the partition backed up? Why amanda is so slow? Is it related to the number of i-nodes or the size of the partition? Thanks for any help Zoltan __ http://www.inf.u-szeged.hu/~kato/ -- research in computer vision http://www.cameradigita.com/ -- photography (online gallery) __
Re: /etc/exclude.gtar is ignored!
On Tue, Nov 11, 2003 at 08:25:31AM +0200, Turgut Kalfaoglu wrote: Hi there. I backup a system, and I notice that it complains about 'unusual or strange backup result' during the email. These are files I forgot to exclude; things like logs, that change during the backup. So, I added them to /etc/exclude.gtar which is defined properly in amanda.conf in 'server-tar', the exclude.gtar is stored at the client, and I am running the client as root (so that it has access to everything), and amcheck shows no problems. yet, every day I keep getting complaints about the same files; like ./apache/cgi-bin/openwebmail/etc/sessions/blah has changed. well, in exclude.gtar I have: ./apache/cgi-bin/openwebmail/etc/sessions and I assumed that this should have stopped everything under 'sessions' from being backed up? I checked the wording, in fact, I did cut/paste from the error emails. I'm obviously missing something! Can anyone help? Questions and a possible theory. 1. Is the file /etc/exclude.gtar mentioned in your amanda.conf with an 'exclude list' or an 'exclude file'? 2. Are any files in /etc/exclude.gtar being excluded? 3. Might your session/* files still being excluded despite the strange messages? I.e. are they listed in the index files? My reason for question 3 relates to my theory. Not knowing the gnutar code, I wonder if it might first notice the change in the file, print its stderr message, and later realize it was a file to be excluded and not backup the file anyway? Just musing! -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: Samba issues
I submitted a bug to samba bugzilla. regards, gregor
Re: can't backup a disk?
On Tuesday 11 November 2003 10:40, Frank Smith wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail Notice that /var/spool/mail is a seperate partion than /var. Dump and tar (with the options Amanda calls it with) does not cross filesytems, so a backup of /var on your machine will not backup /var/spool/mail. Frank I didn't give that much thought because this list looks almost bogus. I've tried to do similar operations to expand disk space here while figuring out howto get the data moved to a larger partition, and linux seems to be able to get fooled in this regard, so I would have had to mount the second /var as /var2 or some such work around. And I need to make a transition to yet another bigger disk here as I have 2Gb /root at 99% its already been moved from hda to hdd when 1Gb was a bottleneck. 120Gb drive wanted, cheap. :) before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved. -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: can't backup a disk?
Thanks again, Gene and Frank! Yeah, my tar is fine, it's 1.13-25. I installed amanda from rpms. So according to Frank, since /var and /var/spool/mail are on different diskes, tar shouldn't cross them, right? Best Chen No, by dump, I meant the program 'dump', as opposed to the program 'tar' (usually gnutar, and it needs to be at least version 1.13-19, plain old 1.13 won't do, its busted, get 1.13-25 from alpha.gnu.org's ftp site) tar should be able to do this, no problem, although if you leave the email agent running full time (I do, I'm on dsl) then its every so often scans of your mailbox at your ISP might occasionally cause a file changed while we read it message from tar if its updateing the mail spool while tar is reading it. However, since you have /var in the disklist, and tar traverses all subdirs below the named one, there must be a more fundamental problem, possibly related to permissions. You did configure and build amanda as the user 'amanda' didn't you? And then become root before installing her? On Tue, 11 Nov 2003, Gene Heskett wrote: On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- === Yu Chen Howard Hughes Medical Institute Chemistry Building, Rm 182 University of Maryland at Baltimore County 1000 Hilltop Circle Baltimore, MD 21250 phone: (410)455-6347 (410)455-2718 fax:(410)455-1174 email: [EMAIL PROTECTED] ===
Re: backup lasts forever on large fs
Hi, Zoltan, on Dienstag, 11. November 2003 at 17:22 you wrote to amanda-users: ZK I'm trying to backup a home partition with amanda under Sun Solaris 9 with ZK Amanda version 2.4.4p1: ZK Filesystem size used avail capacity Mounted on ZK /dev/md/dsk/d0 102G50G51G50%/home ZK which contains a lot of i-nodes: ZK Filesystem iused ifree %iused Mounted on ZK /dev/md/dsk/d0 2932711 992348123% /home ZK I can write it to tape in ~5-6 hours using plain gtar but amanda seems to ZK be busy for several hours (10+) without writing anything to tape. Is there ZK a workaround to get the partition backed up? Why amanda is so slow? Is it ZK related to the number of i-nodes or the size of the partition? I don´t know why you think of i-nodes in this case. Please tell us more about your amanda.conf, disklist and stuff. Also tell us if your Amanda-setup works for other disks. Do you use a holdingdisk? More infos will help us to help you. -- best regards, Stefan Stefan G. Weichinger mailto:[EMAIL PROTECTED]
Re: includefile directive?
On Tue, Nov 11, 2003 at 04:37:17PM +0100, Toralf Lund wrote: Is anyone here using the includefile directive in their config? How exactly does it work? Does it apply to all config files, or just amanda.conf? What can the file contain - full config info, or just whatever is not set in the file including it? If I have two configs, can I have one amanda.conf include the other and override some of the params, or is it better to let both include a 3rd file (or several different files for the different sections)? Under my /usr/local/etc/amanda I have several config directories. I also have a directory called CommonConfFiles. I have stripped down my amanda.conf files to eliminate things like dumptypes, tapetypes, and a few others. Each is in a separate file under CommonConfFiles. Then at the end of my amanda.conf's I use includefile to bring in each file I need. It takes amand.conf from a scary 560 lines to a still imposing 100 lines. jl -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: can't backup a disk?
--On Tuesday, November 11, 2003 11:38:46 -0500 Yu Chen [EMAIL PROTECTED] wrote: Thanks again, Gene and Frank! Yeah, my tar is fine, it's 1.13-25. I installed amanda from rpms. So according to Frank, since /var and /var/spool/mail are on different diskes, tar shouldn't cross them, right? Since Amanda calls tar with the --one-file-system option, it won't follow links across partition boundaries. Dump is restriceted to a single partition as well. It's something you have to keep in mind with Aamnda, a single DLE of / won't backup your entire machine if you have multiple disks and/or partitions. Frank Best Chen No, by dump, I meant the program 'dump', as opposed to the program 'tar' (usually gnutar, and it needs to be at least version 1.13-19, plain old 1.13 won't do, its busted, get 1.13-25 from alpha.gnu.org's ftp site) tar should be able to do this, no problem, although if you leave the email agent running full time (I do, I'm on dsl) then its every so often scans of your mailbox at your ISP might occasionally cause a file changed while we read it message from tar if its updateing the mail spool while tar is reading it. However, since you have /var in the disklist, and tar traverses all subdirs below the named one, there must be a more fundamental problem, possibly related to permissions. You did configure and build amanda as the user 'amanda' didn't you? And then become root before installing her? On Tue, 11 Nov 2003, Gene Heskett wrote: On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- === Yu Chen Howard Hughes Medical Institute Chemistry Building, Rm 182 University of Maryland at Baltimore County 1000 Hilltop Circle Baltimore, MD 21250 phone:(410)455-6347 (410)455-2718 fax: (410)455-1174 email:[EMAIL PROTECTED] === -- Frank Smith [EMAIL PROTECTED] Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
Re: can't backup a disk?
Thanks Frank and Gene, I think I got it. Very helpful list! Chen -- === Yu Chen Howard Hughes Medical Institute Chemistry Building, Rm 182 University of Maryland at Baltimore County 1000 Hilltop Circle Baltimore, MD 21250 phone: (410)455-6347 (410)455-2718 fax:(410)455-1174 email: [EMAIL PROTECTED] ===
Re: can't backup a disk?
On Tuesday 11 November 2003 11:38, Yu Chen wrote: Thanks again, Gene and Frank! Yeah, my tar is fine, it's 1.13-25. I installed amanda from rpms. So according to Frank, since /var and /var/spool/mail are on different diskes, tar shouldn't cross them, right? Best Chen Thats correct, tar stays on the filesystem its started on for each invocation. Installing amanda from rpms can at times be a problem child, and is why the majority of us use the tarballs. No, by dump, I meant the program 'dump', as opposed to the program 'tar' (usually gnutar, and it needs to be at least version 1.13-19, plain old 1.13 won't do, its busted, get 1.13-25 from alpha.gnu.org's ftp site) tar should be able to do this, no problem, although if you leave the email agent running full time (I do, I'm on dsl) then its every so often scans of your mailbox at your ISP might occasionally cause a file changed while we read it message from tar if its updateing the mail spool while tar is reading it. However, since you have /var in the disklist, and tar traverses all subdirs below the named one, there must be a more fundamental problem, possibly related to permissions. You did configure and build amanda as the user 'amanda' didn't you? And then become root before installing her? On Tue, 11 Nov 2003, Gene Heskett wrote: On Tuesday 11 November 2003 09:42, Yu Chen wrote: Hello, I have a client with diskes like this: /dev/hda1 /boot /dev/hda2 / /dev/hda3 /var /dev/hda5 /var/spool/mail before, the amanda works fine with the first three diskes, recently, I added /var/spool/mail to the disklist, but amanda failed to back it up, gave out message: FAILURE AND STRANGE DUMP SUMMARY: hostname /var/spool/mail lev 0 FAILED [disk /var/spool/mail offline?] I don't know where to look into, thanks for any advices! Chen Are you using tar, which can do a subdir, or dump, which does the whole partition or nothing? -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: backup lasts forever on large fs
My config is the following: Amanda server runs on the same machine (i.e. the tape drive, which is a HP DLT 40/80GB attached directly to this server). I can succesfully backup other parttitions from the same machine as well as from other clients. I can also backup a single home directory (I've tested with /home/kato) from the /home partition succesfully. However, when the full /home directory is listed in disklist, the backup runs forever (well, we have stopped it after 10+ hours) and no partitions are backed up (I guess Amanda waits for all partitions before starting any backup). I've mentioned the number of i-nodes as the size of the partition is not that big but we have about 3000 users so the 50 GB is made up of many small files. Hence if anything is special about this partition then it is the number of files. I attach the amanda.conf and disklist files to my email. Thanks again for any help, Zoltan __ http://www.inf.u-szeged.hu/~kato/ -- research in computer vision http://www.cameradigita.com/ -- photography (online gallery) __ On Tue, 11 Nov 2003, Stefan G. Weichinger wrote: Hi, Zoltan, on Dienstag, 11. November 2003 at 17:22 you wrote to amanda-users: ZK I'm trying to backup a home partition with amanda under Sun Solaris 9 with ZK Amanda version 2.4.4p1: ZK Filesystem size used avail capacity Mounted on ZK /dev/md/dsk/d0 102G50G51G50%/home ZK which contains a lot of i-nodes: ZK Filesystem iused ifree %iused Mounted on ZK /dev/md/dsk/d0 2932711 992348123% /home ZK I can write it to tape in ~5-6 hours using plain gtar but amanda seems to ZK be busy for several hours (10+) without writing anything to tape. Is there ZK a workaround to get the partition backed up? Why amanda is so slow? Is it ZK related to the number of i-nodes or the size of the partition? I dont know why you think of i-nodes in this case. Please tell us more about your amanda.conf, disklist and stuff. Also tell us if your Amanda-setup works for other disks. Do you use a holdingdisk? More infos will help us to help you. -- best regards, Stefan Stefan G. Weichinger mailto:[EMAIL PROTECTED] # # amanda.conf - sample Amanda configuration file. This started off life as # the actual config file in use at CS.UMD.EDU. # # If your configuration is called, say, csd, then this file normally goes # in /etc/amanda/csd/amanda.conf. # org cab # your organization name for reports mailto [EMAIL PROTECTED] [EMAIL PROTECTED]# space separated list of operators at your site dumpuser amanda # the user to run dumps under inparallel 4# maximum dumpers that will run in parallel netusage 5 Kbps# maximum net bandwidth for Amanda, in KB per sec dumpcycle 1 weeks # the number of days in the normal dump cycle runspercycle 6 # the number of amdump runs in dumpcycle days tapecycle 7 tapes # the number of tapes in rotation # 4 weeks (dumpcycle) times 5 tapes per week (just # the weekdays) plus a few to handle errors that # need amflush and so we do not overwrite the full # backups performed at the beginning of the previous # cycle ### ### ### # WARNING: don't use `inf' for tapecycle, it's broken! ### ### ### bumpsize 20 Mb # minimum savings (threshold) to bump level 1 - 2 bumpdays 1 # minimum days at each level bumpmult 4 # threshold = bumpsize * bumpmult^(level-1) etimeout 300# number of seconds per filesystem for estimates. #etimeout -600 # total number of seconds for estimates. # a positive number will be multiplied by the number of filesystems on # each host; a negative number will be taken as an absolute total time-out. # The default is 5 minutes per filesystem. # Specify tape device and/or tape changer. If you don't have a tape # changer, and you don't want to use more than one tape per run of # amdump, just comment out the definition of tpchanger. # Some tape changers require tapedev to be defined; others will use # their own tape device selection mechanism. Some use a separate tape # changer device (changerdev), others will simply ignore this # parameter. Some rely on a configuration file (changerfile) to # obtain more information about tape devices, number of slots, etc; # others just need to store some data in files, whose names will start # with changerfile. For more information about individual tape # changers, read docs/TAPE.CHANGERS. # At most one changerfile entry must be defined; select the most # appropriate one for your configuration. If you select man-changer, # keep the first one; if you decide not to use a
Re: can't backup a disk?
On Tue, Nov 11, 2003 at 10:54:03AM -0600, Frank Smith wrote:
--On Tuesday, November 11, 2003 11:38:46 -0500 Yu Chen [EMAIL PROTECTED] wrote:
Since Amanda calls tar with the --one-file-system option, it won't
follow links across partition boundaries. Dump is restriceted to a
single partition as well.
It's something you have to keep in mind with Aamnda, a single
DLE of / won't backup your entire machine if you have multiple disks
and/or partitions.
Something I played with a while ago might allow that.
I found that a single disklist entry could include
mountpoints under the top-level directory. It would
be something like this:
butch.jgcomp.com wholesystem / {
other disklist options
program GNUTAR
include append /usr
include append /export
include append /var
include append /var/spool/mail
include append .
}
I never tried the last entry, something to include the root
file system. But it would be needed to get the entire system.
I only explored whether multiple file systems could get into
a single DLE.
One thing I was considering it for was backing up a project
that might have pieces spread over several file systems.
So you might do something like:
host project / {
...
include append /opt/project
include append /usr/local/project
include append /etc/opt/project
include append /home/jon/Projects/projects/reports
}
Sorry if some of the syntax is not accurate, I don't have
a working example available at this time.
--
Jon H. LaBadie [EMAIL PROTECTED]
JG Computing
4455 Province Line Road(609) 252-0159
Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: backup lasts forever on large fs
ZK My config is the following:
ZK Amanda server runs on the same machine (i.e. the tape drive, which is a HP
ZK DLT 40/80GB attached directly to this server).
Which one? I assume that your machines are the following disklist
entries, even if some are commented out right now.
From your former post I assume rozi.cab.u-szeged.hu is your Amanda
tape server.
Ooops, the disklist file has been changed since I last tried to backup
/home. Of course, the right entry was the following:
rozi.cab.u-szeged.hu/home user-tar
Is the holdingdisk the same physical disk as the on /home is on?
No, it is on another partition (/opt) and partly on the same physical
disk. Here are the relevant parts:
/dev/dsk/c1t0d0s5 20G 1.0G18G 6%/opt
/dev/md/dsk/d0 102G51G51G51%/home
Since /home is under lvm, it uses two disks. Out of the 102G, we have 30G
on the same disk as /opt. However, it is hard to tell how much of the 50G
is actually stored on the same disk as /opt. Anyway, the holding disk is
only 18G so it is not used for the backup of /home.
#disklist
rozi.cab.u-szeged.hu /home {
user-tar
holdingdisk no
}
or creating a specific dumptype which disables holdingdisk.
Disabling this would show if Amanda then is able to dump /home.
OK, I'm going start it right now with these settings,
Regards,
Zoltan
Re: backup lasts forever on large fs
Hi, Zoltan Kato, on Dienstag, 11. November 2003 at 18:51 you wrote to amanda-users: Is the holdingdisk the same physical disk as the on /home is on? ZK No, it is on another partition (/opt) and partly on the same physical ZK disk. Here are the relevant parts: ZK /dev/dsk/c1t0d0s5 20G 1.0G18G 6%/opt ZK /dev/md/dsk/d0 102G51G51G51%/home ZK Since /home is under lvm, it uses two disks. Out of the 102G, we have 30G ZK on the same disk as /opt. However, it is hard to tell how much of the 50G ZK is actually stored on the same disk as /opt. Anyway, the holding disk is ZK only 18G so it is not used for the backup of /home. The files you showed me tell Amanda to use the holdingdisk as it is default and you don´t explicitly specify the holdingdisk parameter anywhere. ZK OK, I'm going start it right now with these settings, Let us know. And look at it with amstatus ... -- best regards, Stefan Stefan G. Weichinger mailto:[EMAIL PROTECTED]
Re: backup lasts forever on large fs
The backup is running now. Here is what top says:
load averages: 0.46, 0.31, 0.18
19:00:37
134 processes: 131 sleeping, 1 zombie, 2 on cpu
CPU states: 16.3% idle, 12.4% user, 2.7% kernel, 68.6% iowait, 0.0% swap
Memory: 2048M real, 1457M free, 190M swap in use, 5385M swap free
PID USERNAME THR PRI NICE SIZE RES STATETIMECPU COMMAND
24118 root 1 100 12M 12M cpu/12:27 22.34% gtar
68% iowait while the only activity on the machine is gtar ran by amanda.
Here is what amstat says:
[EMAIL PROTECTED] amstatus cab
Using /var/amanda/cab/amdump from Tue Nov 11 18:51:59 CET 2003
lena2.cab.u-szeged.hu:/etc 00k estimate done
lena2.cab.u-szeged.hu:/root/ldap_backup 0 1380k estimate done
rozi.cab.u-szeged.hu:/etc getting estimate
rozi.cab.u-szeged.hu:/home getting estimate
SUMMARY part real estimated
size size
partition : 4
estimated : 212490k
flush : 0 0k
failed : 00k ( 0.00%)
wait for dumping: 00k ( 0.00%)
dumping to tape : 00k ( 0.00%)
dumping : 0 0k 0k ( 0.00%) ( 0.00%)
dumped : 0 0k 0k ( 0.00%) ( 0.00%)
wait for writing: 0 0k 0k ( 0.00%) ( 0.00%)
wait to flush : 0 0k 0k (100.00%) ( 0.00%)
writing to tape : 0 0k 0k ( 0.00%) ( 0.00%)
failed to tape : 0 0k 0k ( 0.00%) ( 0.00%)
taped : 0 0k 0k ( 0.00%) ( 0.00%)
all dumpers active
taper idle
Looks like gtar is quite busy in doing the estimates
Zoltan
__
http://www.inf.u-szeged.hu/~kato/ -- research in computer vision
http://www.cameradigita.com/ -- photography (online gallery)
__
On Tue, 11 Nov 2003, Zoltan Kato wrote:
ZK My config is the following:
ZK Amanda server runs on the same machine (i.e. the tape drive, which is a HP
ZK DLT 40/80GB attached directly to this server).
Which one? I assume that your machines are the following disklist
entries, even if some are commented out right now.
From your former post I assume rozi.cab.u-szeged.hu is your Amanda
tape server.
Ooops, the disklist file has been changed since I last tried to backup
/home. Of course, the right entry was the following:
rozi.cab.u-szeged.hu/home user-tar
Is the holdingdisk the same physical disk as the on /home is on?
No, it is on another partition (/opt) and partly on the same physical
disk. Here are the relevant parts:
/dev/dsk/c1t0d0s5 20G 1.0G18G 6%/opt
/dev/md/dsk/d0 102G51G51G51%/home
Since /home is under lvm, it uses two disks. Out of the 102G, we have 30G
on the same disk as /opt. However, it is hard to tell how much of the 50G
is actually stored on the same disk as /opt. Anyway, the holding disk is
only 18G so it is not used for the backup of /home.
#disklist
rozi.cab.u-szeged.hu /home {
user-tar
holdingdisk no
}
or creating a specific dumptype which disables holdingdisk.
Disabling this would show if Amanda then is able to dump /home.
OK, I'm going start it right now with these settings,
Regards,
Zoltan
RE: Problem configuring amanda before compiling
Backups to the file: device ran just fine. The vtblc warnings (QIC support) apparently don't affect anything here. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dana Bourgeois Sent: Monday, November 10, 2003 9:36 PM To: 'Dana Bourgeois'; [EMAIL PROTECTED] Subject: RE: Problem configuring amanda before compiling Answered my own question once I had time to dig a little. You need to have the readline-devel rpm installed as well as the readline rpm. No one came forward about the QIC tape warning but I will know after I configure the disk 'tapes' and run an amdump if it's a problem. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dana Bourgeois Sent: Monday, November 10, 2003 6:44 PM To: [EMAIL PROTECTED] Subject: Problem configuring amanda before compiling On RedHat, I didn't run into this. Now I'm on SsSE8.1 and the ./configure gives me a warning that I don't have a terminal library for interactive amrecover nor do I have the vtblc stuff needed for QIC tape support. Possibly I don't care. I'm going to use the file: driver with this installation and the chg-multi tape changer. I have the readline package installed. Anyone know if I care about these warnings and if so, which packages I need to tickle YaST to install for me? Please believe that I have already tried searching by package name and installed several packages that looked pertinent but didn't make these warnings go away, (ummm... I just reran ./configure - I didn't see anything that would remove/invalidate/unconfigure the previous ./configure step). Clues gratefully accepted. Dana Bourgeois
RE: Problem configuring amanda before compiling
Gene, no QIC drive and amcheck ran fine. Thank you for responding on this point. Re: your previous post about making and installing Amanda. I know the rule of thumb is to 'make install' as root but the Amanda INSTALL file doesn't say that. I think that should be fixed and I'll be happy to send in a proposed change if someone will tell me where it should go. Second thing is that if one does run 'make install' non-root, it runs just fine but the installer can't change the owner which screws up the SUID stuff. It wouldn't be too hard to fix this either and again, I'll be happy to forward a proposed change. Probably just a check that the EUID is root before running the first command. Lastly, a SUID script that fixes up the permissions if they get hosed would be a nice addition to the package and in fact, there is probabaly a lot of other sanity checking and correction that could be bundled into it over time. If this idea has some backing, I'll take a shot at this one also. Basically I'd pull the stuff from the 'make install' into a separate script that could be put into 'libexec'. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gene Heskett Sent: Tuesday, November 11, 2003 2:43 AM To: Dana Bourgeois; [EMAIL PROTECTED] Subject: Re: Problem configuring amanda before compiling On Tuesday 11 November 2003 00:36, Dana Bourgeois wrote: Answered my own question once I had time to dig a little. You need to have the readline-devel rpm installed as well as the readline rpm. No one came forward about the QIC tape warning but I will know after I configure the disk 'tapes' and run an amdump if it's a problem. If you don't actually have a QIC drive, they can be ignored. Its an option to turn on when compiling a new kernel IIRC. The problem finder isn't amdump per sei, but amcheck, which will scan the system and check the install, reporting warnings (which can be ignoored, and errors that will stop the amdump. Errors must be fixed. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dana Bourgeois Sent: Monday, November 10, 2003 6:44 PM To: [EMAIL PROTECTED] Subject: Problem configuring amanda before compiling On RedHat, I didn't run into this. Now I'm on SsSE8.1 and the ./configure gives me a warning that I don't have a terminal library for interactive amrecover nor do I have the vtblc stuff needed for QIC tape support. Possibly I don't care. I'm going to use the file: driver with this installation and the chg-multi tape changer. I have the readline package installed. Anyone know if I care about these warnings and if so, which packages I need to tickle YaST to install for me? Please believe that I have already tried searching by package name and installed several packages that looked pertinent but didn't make these warnings go away, (ummm... I just reran ./configure - I didn't see anything that would remove/invalidate/unconfigure the previous ./configure step). Clues gratefully accepted. Dana Bourgeois -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Answering questions about Amanda on RedHat 7.0
I recently posted a question about my attempts to get 2.4.4p1 running on a RedHat 7.0 box. I probably didn't mention that the plan was to go tapeless. This project didn't work and I can explain exactly why so others won't bump into the same wall I did. 2.4.4p1 compiles just fine on RedHat 7.0 which is a 2.2 kernel. This kernel doesn't have big file support. The file: device also works great. The problem is that while you can create large *tapes* on disk, you can't write large *tape files* on the tape since it is created in a file system that has a 2G limit. I actually ran into this with the holding disk but used the chunksize workaround to get past it without giving thought to what that meant later in the process when taper was running I guess I was too close to the problem and the errors pointed in lots of other places. It wasn't until a coworker was asking about the problems and I mentioned a list of DLEs with the problem and thinking of the list mused ...in fact, almost everything larger than about 2 G is giving an error of some type. This person suggested I run a test to see if a 2.2G file could be created on the host and it could not which made the problem rather obvious. Turns out that the problems all went away with SuSE 8.1 (kernel 2.4.19). In addition I can also verify that this same kernel has built in support for the Promise Ultra 133 TX2 card and will support a 200G IDE drive (Maxtor Diamond Max +9). Dana Bourgeois
Re: Answering questions about Amanda on RedHat 7.0
I recently posted a question about my attempts to get 2.4.4p1 running on a RedHat 7.0 box. I probably didn't mention that the plan was to go tapeless. This project didn't work and I can explain exactly why so others won't bump into the same wall I did. 2.4.4p1 compiles just fine on RedHat 7.0 which is a 2.2 kernel. This kernel doesn't have big file support. The file: device also works great. The problem is that while you can create large *tapes* on disk, you can't write large *tape files* on the tape since it is created in a file system that has a 2G limit. I actually ran into this with the holding disk but used the chunksize workaround to get past it without giving thought to what that meant later in the process when taper was running I guess I was too close to the problem and the errors pointed in lots of other places. It wasn't until a coworker was asking about the problems and I mentioned a list of DLEs with the problem and thinking of the list mused ...in fact, almost everything larger than about 2 G is giving an error of some type. This person suggested I run a test to see if a 2.2G file could be created on the host and it could not which made the problem rather obvious. Turns out that the problems all went away with SuSE 8.1 (kernel 2.4.19). In addition I can also verify that this same kernel has built in support for the Promise Ultra 133 TX2 card and will support a 200G IDE drive (Maxtor Diamond Max +9). why didn't you either upgrade the version of RedHat you were using to something less ancient or just upgrade the kernel on that box?
Re: backup lasts forever on large fs
Looks like the estimate has timed out after a 1/2 hour. I do not know why
estimation takes so long. What is more interesting: after amdump has
finished there is still a gtar process running:
load averages: 0.91, 0.90, 0.90
20:48:53
58 processes: 55 sleeping, 1 zombie, 2 on cpu
CPU states: % idle, % user, % kernel, % iowait, % swap
Memory: 2048M real, 1495M free, 104M swap in use, 5474M swap free
PID USERNAME THR PRI NICE SIZE RES STATETIMECPU COMMAND
24118 root 1 200 49M 49M cpu/1 89:38 45.23% gtar
Here are the results of the test run:
[EMAIL PROTECTED] amstatus cab
Using /var/amanda/cab/amdump.1 from Tue Nov 11 18:51:59 CET 2003
lena2.cab.u-szeged.hu:/etc 1 550k finished (19:37:04)
lena2.cab.u-szeged.hu:/root/ldap_backup 1 1380k finished (19:37:19)
rozi.cab.u-szeged.hu:/etc 0 planner: [Estimate timeout from
rozi.c
ab.u-szeged.hu]
rozi.cab.u-szeged.hu:/home 0 planner: [Estimate timeout from
rozi.c
ab.u-szeged.hu]
SUMMARY part real estimated
size size
partition : 4
estimated : 2 1930k
flush : 0 0k
failed : 20k ( 0.00%)
wait for dumping: 00k ( 0.00%)
dumping to tape : 00k ( 0.00%)
dumping : 0 0k 0k ( 0.00%) ( 0.00%)
dumped : 2 1930k 1930k (100.00%) (100.00%)
wait for writing: 0 0k 0k ( 0.00%) ( 0.00%)
wait to flush : 0 0k 0k (100.00%) ( 0.00%)
writing to tape : 0 0k 0k ( 0.00%) ( 0.00%)
failed to tape : 0 0k 0k ( 0.00%) ( 0.00%)
taped : 2 1930k 1930k (100.00%) (100.00%)
4 dumpers idle : not-idle
taper idle
network free kps:101000
holding space : 18864378k (100.00%)
dumper0 busy : 0:00:00 ( 2.52%)
taper busy : 0:00:06 ( 36.07%)
0 dumpers busy : 0:00:17 ( 95.97%) start-wait: 0:00:14 (
80.88%)
not-idle: 0:00:03 (
19.12%)
1 dumper busy : 0:00:00 ( 4.03%)
[EMAIL PROTECTED]
And amreport says:
-- Forwarded message --
Date: Tue, 11 Nov 2003 19:37:27 +0100 (MET)
From: [EMAIL PROTECTED]
These dumps were to tape cab03.
The next tape Amanda expects to use is: cab04.
FAILURE AND STRANGE DUMP SUMMARY:
rozi.cab.u /home lev 0 FAILED [Estimate timeout from rozi.cab.u-szeged.hu]
rozi.cab.u /etc lev 0 FAILED [Estimate timeout from rozi.cab.u-szeged.hu]
STATISTICS:
Total Full Daily
Estimate Time (hrs:min)0:45
Run Time (hrs:min) 0:45
Dump Time (hrs:min)0:00 0:00 0:00
Output Size (meg) 1.90.01.9
Original Size (meg) 1.90.01.9
Avg Compressed Size (%) -- -- --(level:#disks ...)
Filesystems Dumped2 0 2 (1:2)
Avg Dump Rate (k/s) 5316.8-- 5316.8
Tape Time (hrs:min)0:00 0:00 0:00
Tape Size (meg) 1.90.01.9
Tape Used (%) 0.00.00.0 (level:#disks ...)
Filesystems Taped 2 0 2 (1:2)
Avg Tp Write Rate (k/s) 288.2-- 288.2
USAGE BY TAPE:
Label Time Size %Nb
cab03 0:00 1.90.0 2
NOTES:
planner: Adding new disk rozi.cab.u-szeged.hu:/home.
taper: tape cab03 kb 2048 fm 2 [OK]
DUMP SUMMARY:
DUMPER STATSTAPER STATS
HOSTNAME DISKL ORIG-KB OUT-KB COMP% MMM:SS KB/s MMM:SS KB/s
-- -
lena2.cab.u- /etc1 550550 --0:005662.6 0:03 164.2
lena2.cab.u- -dap_backup 11380 1380 --0:005177.7 0:03 412.2
rozi.cab.u-s /etc0 FAILED ---
rozi.cab.u-s /home 0 FAILED ---
(brought to you by Amanda version 2.4.4p1)
Re: Problem configuring amanda before compiling
On Tuesday 11 November 2003 14:09, Dana Bourgeois wrote: Gene, no QIC drive and amcheck ran fine. Thank you for responding on this point. Good. Re: your previous post about making and installing Amanda. I know the rule of thumb is to 'make install' as root but the Amanda INSTALL file doesn't say that. I think that should be fixed and I'll be happy to send in a proposed change if someone will tell me where it should go. Yes, I'm somewhat surprised that this has not made it to the INSTALL file. Second thing is that if one does run 'make install' non-root, it runs just fine but the installer can't change the owner which screws up the SUID stuff. It wouldn't be too hard to fix this either and again, I'll be happy to forward a proposed change. Probably just a check that the EUID is root before running the first command. OTOH, if you become root, one can fix all the owner:group stuff to match the rest of the system in one swell foop, and a rerun of the make install will then fix the rest of it. IMO not a lot of trouble, it can all be fixed in 30 seconds, and it teaches us to do it right the next time... Lastly, a SUID script that fixes up the permissions if they get hosed would be a nice addition to the package and in fact, there is probabaly a lot of other sanity checking and correction that could be bundled into it over time. If this idea has some backing, I'll take a shot at this one also. Basically I'd pull the stuff from the 'make install' into a separate script that could be put into 'libexec'. See above. Dana Bourgeois [...] -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: Problem configuring amanda before compiling
On Tue, Nov 11, 2003 at 11:09:04AM -0800, Dana Bourgeois wrote: Re: your previous post about making and installing Amanda. I know the rule of thumb is to 'make install' as root but the Amanda INSTALL file doesn't say that. I think that should be fixed and I'll be happy to send in a proposed change if someone will tell me where it should go. From INSTALL: A. Back at the top-level source directory, build the sources: 1. make 2. su root; make install ^^^ -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Answering questions about Amanda on RedHat 7.0
On Tuesday 11 November 2003 14:38, Dana Bourgeois wrote: I recently posted a question about my attempts to get 2.4.4p1 running on a RedHat 7.0 box. I probably didn't mention that the plan was to go tapeless. This project didn't work and I can explain exactly why so others won't bump into the same wall I did. 2.4.4p1 compiles just fine on RedHat 7.0 which is a 2.2 kernel. This kernel doesn't have big file support. The file: device also works great. The problem is that while you can create large *tapes* on disk, you can't write large *tape files* on the tape since it is created in a file system that has a 2G limit. I actually ran into this with the holding disk but used the chunksize workaround to get past it without giving thought to what that meant later in the process when taper was running I guess I was too close to the problem and the errors pointed in lots of other places. It wasn't until a coworker was asking about the problems and I mentioned a list of DLEs with the problem and thinking of the list mused ...in fact, almost everything larger than about 2 G is giving an error of some type. This person suggested I run a test to see if a 2.2G file could be created on the host and it could not which made the problem rather obvious. Turns out that the problems all went away with SuSE 8.1 (kernel 2.4.19). In addition I can also verify that this same kernel has built in support for the Promise Ultra 133 TX2 card and will support a 200G IDE drive (Maxtor Diamond Max +9). Well, its been recognised that rh7.0 wasn't their best effort, particularly from the security aspect. 7.3 was much better in that regard. It should be brought up to 8.0, and the kde portion overwritten with the real kde stuffs. Here, I'm running 2.6.0-test9-mm2 on an 8.0 system. Using the elevator=deadline option on the grub command line as the anticipatory scheduler seems a bit busted. Other than that, its a much more responsive system than any 2.4.x kernel was. Dana Bourgeois -- Cheers, Gene AMD [EMAIL PROTECTED] 320M [EMAIL PROTECTED] 512M 99.27% setiathome rank, not too shabby for a WV hillbilly Yahoo.com attornies please note, additions to this message by Gene Heskett are: Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
Re: backup lasts forever on large fs
--On Tuesday, November 11, 2003 20:50:58 +0100 Zoltan Kato [EMAIL PROTECTED] wrote:
Looks like the estimate has timed out after a 1/2 hour. I do not know why
estimation takes so long. What is more interesting: after amdump has
finished there is still a gtar process running:
Estimates can take quite a while on large filsystems with lots of small
files. You can try increasing etimeout in your amanda.conf, the default
was too short for several of my filesystems. Since etimeout tells your
backup server how long to wait before giving up on the estimate, the
client doesn't know that the server quit waiting so it continues on.
If you let it run you can check the debug logs on the client (usually
in /tmp/amanda unless you configured it differently) and see how long
the estimate ran and increase your etimeout accordingly.
Frank
load averages: 0.91, 0.90, 0.90
20:48:53
58 processes: 55 sleeping, 1 zombie, 2 on cpu
CPU states: % idle, % user, % kernel, % iowait, % swap
Memory: 2048M real, 1495M free, 104M swap in use, 5474M swap free
PID USERNAME THR PRI NICE SIZE RES STATETIMECPU COMMAND
24118 root 1 200 49M 49M cpu/1 89:38 45.23% gtar
Here are the results of the test run:
[EMAIL PROTECTED] amstatus cab
Using /var/amanda/cab/amdump.1 from Tue Nov 11 18:51:59 CET 2003
lena2.cab.u-szeged.hu:/etc 1 550k finished (19:37:04)
lena2.cab.u-szeged.hu:/root/ldap_backup 1 1380k finished (19:37:19)
rozi.cab.u-szeged.hu:/etc 0 planner: [Estimate timeout from
rozi.c
ab.u-szeged.hu]
rozi.cab.u-szeged.hu:/home 0 planner: [Estimate timeout from
rozi.c
ab.u-szeged.hu]
SUMMARY part real estimated
size size
partition : 4
estimated : 2 1930k
flush : 0 0k
failed : 20k ( 0.00%)
wait for dumping: 00k ( 0.00%)
dumping to tape : 00k ( 0.00%)
dumping : 0 0k 0k ( 0.00%) ( 0.00%)
dumped : 2 1930k 1930k (100.00%) (100.00%)
wait for writing: 0 0k 0k ( 0.00%) ( 0.00%)
wait to flush : 0 0k 0k (100.00%) ( 0.00%)
writing to tape : 0 0k 0k ( 0.00%) ( 0.00%)
failed to tape : 0 0k 0k ( 0.00%) ( 0.00%)
taped : 2 1930k 1930k (100.00%) (100.00%)
4 dumpers idle : not-idle
taper idle
network free kps:101000
holding space : 18864378k (100.00%)
dumper0 busy : 0:00:00 ( 2.52%)
taper busy : 0:00:06 ( 36.07%)
0 dumpers busy : 0:00:17 ( 95.97%) start-wait: 0:00:14 (
80.88%)
not-idle: 0:00:03 (
19.12%)
1 dumper busy : 0:00:00 ( 4.03%)
[EMAIL PROTECTED]
And amreport says:
-- Forwarded message --
Date: Tue, 11 Nov 2003 19:37:27 +0100 (MET)
From: [EMAIL PROTECTED]
These dumps were to tape cab03.
The next tape Amanda expects to use is: cab04.
FAILURE AND STRANGE DUMP SUMMARY:
rozi.cab.u /home lev 0 FAILED [Estimate timeout from rozi.cab.u-szeged.hu]
rozi.cab.u /etc lev 0 FAILED [Estimate timeout from rozi.cab.u-szeged.hu]
STATISTICS:
Total Full Daily
Estimate Time (hrs:min)0:45
Run Time (hrs:min) 0:45
Dump Time (hrs:min)0:00 0:00 0:00
Output Size (meg) 1.90.01.9
Original Size (meg) 1.90.01.9
Avg Compressed Size (%) -- -- --(level:#disks ...)
Filesystems Dumped2 0 2 (1:2)
Avg Dump Rate (k/s) 5316.8-- 5316.8
Tape Time (hrs:min)0:00 0:00 0:00
Tape Size (meg) 1.90.01.9
Tape Used (%) 0.00.00.0 (level:#disks ...)
Filesystems Taped 2 0 2 (1:2)
Avg Tp Write Rate (k/s) 288.2-- 288.2
USAGE BY TAPE:
Label Time Size %Nb
cab03 0:00 1.90.0 2
NOTES:
planner: Adding new disk rozi.cab.u-szeged.hu:/home.
taper: tape cab03 kb 2048 fm 2 [OK]
DUMP SUMMARY:
DUMPER STATSTAPER STATS
HOSTNAME DISKL ORIG-KB OUT-KB COMP% MMM:SS KB/s MMM:SS KB/s
-- -
lena2.cab.u- /etc1 550550 --0:005662.6 0:03 164.2
lena2.cab.u- -dap_backup 11380 1380 --0:005177.7 0:03 412.2
rozi.cab.u-s /etc0 FAILED ---
rozi.cab.u-s /home 0 FAILED ---
Re: backup lasts forever on large fs
On Tue, Nov 11, 2003 at 08:50:58PM +0100, Zoltan Kato wrote: Looks like the estimate has timed out after a 1/2 hour. I do not know why estimation takes so long. What is more interesting: after amdump has finished there is still a gtar process running: You mentioned, I think, that this is a file system with 2M inodes (files and directories) consumed. FWIW, my experience on Solaris 8 ufs with inode counts around 1M, and GNU tar 1.13.19/1.13.25, is that estimates are very slow. The backup is slow, too, but you only do that once-- estimates (always? typically?) run 3X for level 0 and two levels of incremental. I changed to ufsdump because of this. I assume the --listed-incremental functionality is the problem, but don't know if it is a bug or a feature, or what. Make sure that nothing (or as close to nothing as possible) is competing with amanda for disk arm movement; I had etimeout and dtimeout problems with amanda once, but only on Saturday night; there was a weekly 'find / -name core' type job running at the same time, and the contention for disk resources was killing both jobs. -- Jay Lessert [EMAIL PROTECTED] Accelerant Networks Inc. (voice)1.503.439.3461 Beaverton OR, USA(fax)1.503.466.9472
RE: Answering questions about Amanda on RedHat 7.0
The system is question belongs to an engineer as his personal workstation. I really wanted a stand alone backup server so this solution is my preferred one. I think upgrading the 7.0 system to 8.0 or at least to a 2.4.19 kernel would also have worked just fine. Dana Bourgeois -Original Message- From: Tom Brown [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 11, 2003 11:45 AM To: Dana Bourgeois; [EMAIL PROTECTED] Subject: Re: Answering questions about Amanda on RedHat 7.0 I recently posted a question about my attempts to get 2.4.4p1 running on a RedHat 7.0 box. I probably didn't mention that the plan was to go tapeless. This project didn't work and I can explain exactly why so others won't bump into the same wall I did. 2.4.4p1 compiles just fine on RedHat 7.0 which is a 2.2 kernel. This kernel doesn't have big file support. The file: device also works great. The problem is that while you can create large *tapes* on disk, you can't write large *tape files* on the tape since it is created in a file system that has a 2G limit. I actually ran into this with the holding disk but used the chunksize workaround to get past it without giving thought to what that meant later in the process when taper was running I guess I was too close to the problem and the errors pointed in lots of other places. It wasn't until a coworker was asking about the problems and I mentioned a list of DLEs with the problem and thinking of the list mused ...in fact, almost everything larger than about 2 G is giving an error of some type. This person suggested I run a test to see if a 2.2G file could be created on the host and it could not which made the problem rather obvious. Turns out that the problems all went away with SuSE 8.1 (kernel 2.4.19). In addition I can also verify that this same kernel has built in support for the Promise Ultra 133 TX2 card and will support a 200G IDE drive (Maxtor Diamond Max +9). why didn't you either upgrade the version of RedHat you were using to something less ancient or just upgrade the kernel on that box?
RE: Problem configuring amanda before compiling
I just did a 'grep root INSTALL' in the amanda-2.4.4p1 directory from the tar.gz file I downloaded with the same name from amanda.org and nothing was found. I have to stand by my original comment. Dana Bourgeois -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jon LaBadie Sent: Tuesday, November 11, 2003 12:13 PM To: [EMAIL PROTECTED] Subject: Re: Problem configuring amanda before compiling On Tue, Nov 11, 2003 at 11:09:04AM -0800, Dana Bourgeois wrote: Re: your previous post about making and installing Amanda. I know the rule of thumb is to 'make install' as root but the Amanda INSTALL file doesn't say that. I think that should be fixed and I'll be happy to send in a proposed change if someone will tell me where it should go. From INSTALL: A. Back at the top-level source directory, build the sources: 1. make 2. su root; make install ^^^ -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: Problem configuring amanda before compiling
On Tue, 11 Nov 2003 at 1:09pm, Dana Bourgeois wrote I just did a 'grep root INSTALL' in the amanda-2.4.4p1 directory from the tar.gz file I downloaded with the same name from amanda.org and nothing was found. I have to stand by my original comment. From the top of that file: Basic Installation == These are generic installation instructions. ^^^ That file goes out with every tarball which uses 'configure'. The amanda-specific INSTALL file is in the docs directory. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: backup lasts forever on large fs
Hi, Zoltan Kato, on Dienstag, 11. November 2003 at 20:50 you wrote to amanda-users: ZK Looks like the estimate has timed out after a 1/2 hour. I do not know why ZK estimation takes so long. What is more interesting: after amdump has ZK finished there is still a gtar process running. As Jay and Frank already have recommended, I think increasing the etimeout value is the way to go. According to your posting you have 300 seconds for that right now ... that´s 5 minutes. This is the default value and seems too low for your situation. You have loads of files on that partition, so tar has lots of work to do, as your former posting of the output of top shows. The sendsize.debug files in your /tmp/amanda-directory (or whereever your amanda-installation puts its logfiles ) tell you about the commands Amanda uses for estimating. You can find those commands by looking for lines containing something like getting size via gnutar for /home level 0 Some lines later there follows something like argument list: /bin/tar Run that command manually, maybe even using the time-command: # time tar ... This will run for a while, giving back the time it took. Set your etimeout somewhat higher and try a run again. --- Otherwise set it up to something like 3600 (one hour) and try that. --- Another way would be to split the /home-directory into some smaller disklist entries. But more on that later if necessary. -- best regards, Stefan Stefan G. Weichinger mailto:[EMAIL PROTECTED]
configure bug or misunderstanding?
Hi all! I've been having trouble getting amanda to support gnu tar on Solaris boxes. What I discovered was that even though I explicitly set the path to gnu tar at configure time; ./configure other options --with-gnutar=/usr/local/bin/tar and even though amanda appeared to configure for gnu tar (judging from the config log file and the debugging output where it actually shows the path I gave configure) it refused to recognise it, claiming that it wasn't a gnu tar. Finally I seem to have figured it out; If $PATH has another version of tar which would be found before the gnu tar, even though one explicitly puts it on the configure commandline, the #ifdef's in runtar.c don't appear to pick it up and hence its not compiled in; all that gets compiled in are the messages bitching about not having gnu tar. Or something like that. By setting my $PATH at configure time so that /usr/local/bin was at the front, I got a build on Sloarsis which works with gnu tar! Bug in the configure script? Or am I misunderstanding something? Cheers!
Re: configure bug or misunderstanding?
On Wed, Nov 12, 2003 at 02:20:38PM +1300, Steve Wray wrote: Hi all! I've been having trouble getting amanda to support gnu tar on Solaris boxes. What I discovered was that even though I explicitly set the path to gnu tar at configure time; ./configure other options --with-gnutar=/usr/local/bin/tar and even though amanda appeared to configure for gnu tar (judging from the config log file and the debugging output where it actually shows the path I gave configure) it refused to recognise it, claiming that it wasn't a gnu tar. Finally I seem to have figured it out; If $PATH has another version of tar which would be found before the gnu tar, even though one explicitly puts it on the configure commandline, the #ifdef's in runtar.c don't appear to pick it up and hence its not compiled in; all that gets compiled in are the messages bitching about not having gnu tar. Or something like that. I'm on Solaris. There are 3 tar's, 2 gtar's, and several other *tar's in my path BEFORE the one in the location I use for amanda. A couple after too, my PATH is a ridiculous 74 bin's long. It picks up my --with-guntar just fine. By setting my $PATH at configure time so that /usr/local/bin was at the front, I got a build on Sloarsis which works with gnu tar! Bug in the configure script? Or am I misunderstanding something? Cheers! End of included message -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
http://www.toti.com.tr/kampanya_haziran.htm
http://www.toti.com.tr/kampanya_haziran.htm Adresini Tklatn 100'lerce bilgisayar modeli bir arada Telefonla bilgi iin : (212) 543
