Re: Win32 Backup?
Anthony A. D. Talltree wrote: I've never had to do this, but an approach that I'd probably persue would be to use Symmantec/Norton Ghost to create occasional disk images that would get xferred via Samba to a *ix filesystem. I'd try real hard to keep important data off of the M$ machine's local disks. One way I have done it is to use MS's backup program to backup to a samba share on a freebsd box. MS backup will happily backup to a samba share. Then I just backup all that stuff using amanda. I've used MS backup on XP Pro and w2kpro this way and it works pretty well for my needs. -- Michael Perry | Do or do not. There is no try. -Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
new samba backup errors
I decided to focus on getting smbclient backups working with amanda so read the docs/samba stuff and added relevant entries to my disklist file and also ran amcheck a few times. I am running amanda on a Debian Unstable system with a very recent CVS version of Samba that is in the unstable tree of debian. The tape device is a ecrix vxa-1 which has worked flawlessly backing up bsd and linux systems. The evening backup just concluded and I got the backup report from amanda. The message is filled with errors like this: ? ERROR: string overflow by 10 in safe_strcpy [\Documents and +Settings\mperry\Local Settings\Temp] ? ERROR: string overflow by 18 in safe_strcpy [\Documents and +Settings\mperry\Local Settings\Temp] ? ERROR: string overflow by 16 in safe_strcpy [\Documents and +Settings\mperry\Local Settings\Temp] ? ERROR: string overflow by 20 in safe_strcpy [\Documents and +Settings\mperry\Local Settings\Temp] and several NT sharing violations. Being a newbie at the samba and smbclient backups, I am wondering what these repeated errors mean. Most of them occur in the Documents and Settings folder. The system being backed up is a Windows 2000 pro box. Thanks for any enlightenment. BTW, I set up /etc/amandapass per the readme documentation to backup the C$ share and I am an administrator on that system with an admin login. Amcheck runs with no problems at all. -- Michael Perry [EMAIL PROTECTED]
status of cygwin and amanda
Hi all- I know there is a amanda gui client and applications for win32; but I have a working cygwin build on several systems primarily to use some mail tools like mutt. I also read in the archive that some folks have attempted or perhaps completed a port of the amanda-client to cygwin. I use cygwin on windows 2000 pro here. The cygwin mailing list points to a finished amanda-client port but I cannot find the software anywhere after the initial announcement by the person that did the port. Can anyone direct me to a location for a download of a ported amanda or perhaps give me some information on getting amanda-client working on cygwin? Thanks! -- Michael Perry [EMAIL PROTECTED]
cautionary note about KT133 southbridge systems
HI all- Awhile ago, I built a system which was destined to be my webserver which was a AMD Duron 800 with the fabled Kt133a Southbridge chipset. AFter numerous backups with amanda, I kept on getting i/o errors and strange dump reports using an IBM 45 gig ATA 100 drive. On the system, the drive was filled with dma retries and resets. I tried a whole bunch of stuff to solve this including stepping the system down to no-DMA but still had the problems. This was on a 2.4.13 kernel with the VIA ide driver enabled; which usually fixes this sort of thing. At the end of things, the drive started clanking on copying my build to a replacement disk drive and I realized that the drive was probably gonna fail on me at any moment. It also made me doubt that validity of backups since each one seemed to have the strange dump reports in it but they all completed. My main note here is to move carefully when using this particular chipset and controller and different IDE drives. For me, any amount of intensive disk io caused the system to burp out literally hundreds of dma reset/retry errors. Replacing the drive with a Maxtor DiamondMax 80g drive solved the problems completely. If you see dma resets/retries when doing amanda dumps, you may want to doublecheck what you are using as far as dma settings on the drive. -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
help with amrestore needed
Hi all- I am in the process of testing some recovery using dump versus switching to gnutar and wanted to begin by using amrestore against a tape volume in my tapeserver. The test file is a binary debian package of opera wich was on hdb1 on the desktop. The way I read the man page for amrestore is that I issue a command like this somewhat: amrestore -p /dev/nrmt9 seine rz3g | restore -ivbf 2 - I replaced the /dev/ line with my /dev/nst0 and seine with homebase and then I replaced the rz3g with hdb1 and piped restore as it shows. I also checked to make sure I had the correct entries in my /etc/services file on the client system and I do. As I type the command in, I cannot even access /dev/nst0 on the tapeserver by issuing the command: amrestore -p /dev/nst0 homebase hdb1 | restore -ivbf 2 - The answer I get is simply that the device /dev/nst0 does not exist. I rewound the tape drive using mt -f and it works quite well. Since I felt this could be something around networking and hostnames, I checked my /etc/hosts file and I have an entry for homebase and I can backup to homebase using amdump every other night. amrestore works on the tapeserver quite well. Is there some other file I need to edit like a config file to make amrestore work remotely? I believe I have networking up correctly. Thanks for any help! -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
Clarification of dump in 2.4 kernels
I recently posted on comp.os.linux.setup regarding some non-amanda related issues with dma on drives and I had actually posted here awhile ago as well about these same issues. A respondent told me about several issues with dump and recent 2.4 kernels. I found some issues with specifically the 2.4.4 and 2.4.5 kernels regarding file system buffering with ext2 and recent 2.4.x kernels. I have been using dump for quite some time on two of three systems and don't notice any problems but Torvalds called into question the validity of the backups when using dump and 2.4 kernels. Is this still the case with dump and a kernel like a self-compiled 2.4.17 kernel running with debian unstable with the latest debian packages? I have switched over to using gnutar for most other things but I still use dump on one system which does webserving for me. The last thing I could find after a rather quick search was an article on lwn.net about using dump and an email that Linus Torvalds posted to the kernel mailing list. One of the followups specifically stated that SMP was an issue. Thanks for any clarification or reading assignments :) -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
Re: Clarification of dump in 2.4 kernels
Quoting Joshua Baker-LePain on Tue, Feb 26, 2002 at 02:43:14PM -0500:
On Tue, 26 Feb 2002 at 11:26am, Michael Perry wrote
I have been using dump for quite some time on two of three systems and
don't notice any problems but Torvalds called into question the validity
of the backups when using dump and 2.4 kernels.
Is this still the case with dump and a kernel like a self-compiled
2.4.17 kernel running with debian unstable with the latest debian
packages? I have switched over to using gnutar for most other things
but I still use dump on one system which does webserving for me.
AHHH!
\begin{monty_python}
Run Away! Run Away!
\end{monty_python}
Err, sorry 'bout that. But this little flame war erupted back in the fall
(IIRC), and we don't want to do it again. The upshot is that some people
see random date timeouts on 2.4.x Linux systems using dump, and those
issues go away with tar.
But what you should use is what you are comfortable with and what passes
*your* tests.
--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University
Thanks!
I did not see any of the previous posts on the issue so I did not dig
deeply enough to find the Python :)
I don't want to be accused of restarting any old flamage.
--
Michael Perry | Do or do not; there is no try Master Yoda
[EMAIL PROTECTED] | http://www.lnxpowered.org
cautionary note about KT133 southbridge systems
HI all- Awhile ago, I built a system which was destined to be my webserver which was a AMD Duron 800 with the fabled Kt133a Southbridge chipset. AFter numerous backups with amanda, I kept on getting i/o errors and strange dump reports using an IBM 45 gig ATA 100 drive. On the system, the drive was filled with dma retries and resets. I tried a whole bunch of stuff to solve this including stepping the system down to no-DMA but still had the problems. This was on a 2.4.13 kernel with the VIA ide driver enabled; which usually fixes this sort of thing. At the end of things, the drive started clanking on copying my build to a replacement disk drive and I realized that the drive was probably gonna fail on me at any moment. It also made me doubt that validity of backups since each one seemed to have the strange dump reports in it but they all completed. My main note here is to move carefully when using this particular chipset and controller and different IDE drives. For me, any amount of intensive disk io caused the system to burp out literally hundreds of dma reset/retry errors. Replacing the drive with a Maxtor DiamondMax 80g drive solved the problems completely. If you see dma resets/retries when doing amanda dumps, you may want to doublecheck what you are using as far as dma settings on the drive. -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
tape io errors on Ecrix VXA-1
I am curious if people see failed backups regularly, occasionally, or hardly at all due to IO errors on the tapeserver bus. My own failed backups seem to be linked to questionable tapes which may have been ready to fail anyways. Unfortunately, the ecrix tapes are very expensive to me and I hate having to replace them since each one is almost $100. The tape errors are about rewinding or issues with the tape itself. What are other's experiences with the VXA-1 or is tape just a dirty media and tapes fail very often, etc? I wish I knew the history of some of the tapes, but I don't. I also upgraded the firmware on the VXA-1 since I have no idea when that was last done. I have run a tape cleaner through the tape drive recently (perhaps a week ago or so). Secondarily, I am curious if people have recommendations for scsi-based tape drive solutions which would work well on a home LAN configured as follows: 2 debian unstable systems doing desktop and one webserver on testing 1 Windows XP Pro box (workstation/desktop) 1 BSD box doing critical stuff (firewall) 1 Windows 98 SE box (kidster computer) 1 debian laptop The linux systems are all running a recent kernel. Each system probably has a few gigs of used up space except for the BSD box. Now the VXA-1 backs up all the *nix stuff easily on one tape and I like that operation! Thanks and a happy New Year to all of you! -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
tape io errors on Ecrix VXA-1
I am curious if people see failed backups regularly, occasionally, or hardly at all due to IO errors on the tapeserver bus. My own failed backups seem to be linked to questionable tapes which may have been ready to fail anyways. Unfortunately, the ecrix tapes are very expensive to me and I hate having to replace them since each one is almost $100. The tape errors are about rewinding or issues with the tape itself. What are other's experiences with the VXA-1 or is tape just a dirty media and tapes fail very often, etc? I wish I knew the history of some of the tapes, but I don't. I also upgraded the firmware on the VXA-1 since I have no idea when that was last done. I have run a tape cleaner through the tape drive recently (perhaps a week ago or so). Secondarily, I am curious if people have recommendations for scsi-based tape drive solutions which would work well on a home LAN configured as follows: 2 debian unstable systems doing desktop and one webserver on testing 1 Windows XP Pro box (workstation/desktop) 1 BSD box doing critical stuff (firewall) 1 Windows 98 SE box (kidster computer) 1 debian laptop The linux systems are all running a recent kernel. Each system probably has a few gigs of used up space except for the BSD box. Now the VXA-1 backs up all the *nix stuff easily on one tape and I like that operation! Thanks and a happy New Year to all of you! -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
cautionary note about KT133 southbridge systems
HI all- Awhile ago, I built a system which was destined to be my webserver which was a AMD Duron 800 with the fabled Kt133a Southbridge chipset. AFter numerous backups with amanda, I kept on getting i/o errors and strange dump reports using an IBM 45 gig ATA 100 drive. On the system, the drive was filled with dma retries and resets. I tried a whole bunch of stuff to solve this including stepping the system down to no-DMA but still had the problems. This was on a 2.4.13 kernel with the VIA ide driver enabled; which usually fixes this sort of thing. At the end of things, the drive started clanking on copying my build to a replacement disk drive and I realized that the drive was probably gonna fail on me at any moment. It also made me doubt that validity of backups since each one seemed to have the strange dump reports in it but they all completed. My main note here is to move carefully when using this particular chipset and controller and different IDE drives. For me, any amount of intensive disk io caused the system to burp out literally hundreds of dma reset/retry errors. Replacing the drive with a Maxtor DiamondMax 80g drive solved the problems completely. If you see dma resets/retries when doing amanda dumps, you may want to doublecheck what you are using as far as dma settings on the drive. -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
replacing full tapes?
I have been busily backing up my home network which is about 5 linux/bsd systems onto an ecrix vxa tape drive. I finally got these errors this evening on a bup tape: *** A TAPE ERROR OCCURRED: [cannot overwrite active tape BUP2]. Some dumps may have been left in the holding disk. Run amflush to flush them to tape. The next tape Amanda expects to use is: BUP3. FAILURE AND STRANGE DUMP SUMMARY: wwwhda2 lev 0 FAILED [disk hda2 offline on www?] wwwhda1 lev 0 FAILED [disk hda1 offline on www?] mperry hda2 lev 2 FAILED [no more holding disk space] I understand the issue with www. I did not install dump on this system before running the backup. How do I correct the tape error above? As far as holding disk space goes, I have a whole bunch left and amcheck reports that I have over 35g of space left which is plenty. I now have cycled to bup3 which is next in line for backup. I corrected the issue with 'www and the lack of dump on that system. I don't quite understand how to deal with the tape error about overwriting active tape bup2 above. Thanks. -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
Re: replacing full tapes?
Quoting Bort, Paul on Wed, Dec 12, 2001 at 10:49:31AM -0500: You probably don't want to overwrite BUP2. IIRC, active tape means that the most recent combination of a specific disk and level exists on that tape. The amoverview command can show what levels were backed up on which days, for starters. You might also want to look at your tapecycle, dumpcycle, and runspercycle parameters. I usually try to keep dumpcycle at half (or less) of tapecycle, so that I always have a complete set of tapes to restore from, and the set I'm currently overwriting. Hi- Thanks for the pointer. I had neglected to take into account a few factors about using the settings like dumpcycle, tapecycle, runspercycle. My basic need I think is to backup 5 systems 3 times per week and have a 2 week latency of backups. This is not a huge network :) but I really enjoy the way amanda does the scheduling, planning, etc. Best I have ever found in some years of Linux work. What I am considering is how to build the correct settings for this and not have to buy a lot of very expensive ecrix tapes (almost $100 per tape). I have sufficient tapes to handle up to a 14 tape tapecycle but some of them are rather old I think and many are new so I am unsure about the relative health of the tapes. If I set the following variables, does this look close to my goal? BTW, the goal is to backup systems 3 times per week (Mon, Wed, Fri) and have two weeks of backups: dumpcycle 2 weeks tapecycle 8 tapes runspercycle 3 I am assuming with this setup that amanda will backup 3 times per week according to my crontab and check 3 times per week a bit earlier for the right tape. If I want 2 weeks of backups this would equal 6 tapes and then I would toss in a few extra to take care of any issues like failures, etc. I am backing up 5 systems here at home including a bsd firewall box which is exceedingly important to my sanity. Perhaps the other question is how other smaller LAN users schedule weekly backups or two week backups when doing dumps not every day but perhaps every other day and excluding weekends. Thanks for the answers to the questions everyone has sent. I should have provided all the information on my goals and approaches using amanda first time around. -- Michael Perry | Do or do not; there is no try Master Yoda [EMAIL PROTECTED] | http://www.lnxpowered.org
