Re: speed problem
On Thu, 6 Feb 2003 [EMAIL PROTECTED] wrote: > My tape server is a Linux host (Redhat 7.3). > My clients are either Linux hosts or Sun Solaris hosts. > All client hosts are on the same domain. > When I run 'amdump' to backup Linux hosts, the speed is pretty good. > When I run 'amdump' to backup Sun Solaris hosts, the speed is extremely > slow. > It seems not a network or hardware issue. Seems? You don't sound particularly sure on this point... In my experience one thing that is a frequent cause of deadly slow network backups is a duplex mismatch between a backup client and the ethernet switch it connects to. I've seen systems that the user had been happily using without complaint for weeks or months before asking for backups. When their backup went painfully slow I'd check their network settings and invariably find a duplex mismatch. Another cause of slowness can be doing client compression, especially "client best" on slow hardware. How old/slow are your Solaris boxen? -Mitch
speed problem
My tape server is a Linux host (Redhat 7.3). My clients are either Linux hosts or Sun Solaris hosts. All client hosts are on the same domain. When I run 'amdump' to backup Linux hosts, the speed is pretty good. When I run 'amdump' to backup Sun Solaris hosts, the speed is extremely slow. It seems not a network or hardware issue. The 'amdump' use gnutar utility. We installed gnutar version 1.13-25. Is any suggestion? Thanks! David
Re: amdump problem
On Thu, Feb 06, 2003 at 06:46:37PM -0300, Luis E. Cruz Campos enlightened us: > I have a problem with my server, i try configure amanda, buy I get this mail > when run amdump: > > These dumps were to tape hosting01. > The next tape Amanda expects to use is: a new tape. > > FAILURE AND STRANGE DUMP SUMMARY: > ns.webspac /usr/local/backup/data lev 0 FAILED [/sbin/dump returned 1] > *snip* > | DUMP: You can't update the dumpdates file when dumping a subdirectory > | DUMP: The ENTIRE dump is aborted. > sendbackup: error [/sbin/dump returned 1] > I'd say this is your error. My guess is that /usr/local/backup/data is not its own partition, and as a result you can't use dump. I would suggest changing to tar if you just want the subdirectory. Matt -- Matt Hyclak Department of Mathematics Ohio University (740) 593-1263 msg17537/pgp0.pgp Description: PGP signature
Re: amdump problem
On many (some?) OSs you can't use dump on a subdirectory. Use GNU tar
instead, or use dump on the entire filesystem.
Frank
--On Thursday, February 06, 2003 18:46:37 -0300 "Luis E. Cruz Campos" <[EMAIL PROTECTED]> wrote:
I have a problem with my server, i try configure amanda, buy I get this mail
when run amdump:
These dumps were to tape hosting01.
The next tape Amanda expects to use is: a new tape.
FAILURE AND STRANGE DUMP SUMMARY:
ns.webspac /usr/local/backup/data lev 0 FAILED [/sbin/dump returned 1]
STATISTICS:
Total Full Daily
Estimate Time (hrs:min)0:00
Run Time (hrs:min) 0:01
Dump Time (hrs:min)0:00 0:00 0:00
Output Size (meg) 0.00.00.0
Original Size (meg) 0.00.00.0
Avg Compressed Size (%) -- -- --
Filesystems Dumped0 0 0
Avg Dump Rate (k/s) -- -- --
Tape Time (hrs:min)0:00 0:00 0:00
Tape Size (meg) 0.00.00.0
Tape Used (%) 0.00.00.0
Filesystems Taped 0 0 0
Avg Tp Write Rate (k/s) -- -- --
FAILED AND STRANGE DUMP DETAILS:
/-- ns.webspac /usr/local/backup/data lev 0 FAILED [/sbin/dump returned 1]
sendbackup: start [ns.xxx.com:/usr/local/backup/data level 0]
sendbackup: info BACKUP=/sbin/dump
sendbackup: info RECOVER_CMD=/sbin/restore -f... -
sendbackup: info end
| DUMP: You can't update the dumpdates file when dumping a subdirectory
| DUMP: The ENTIRE dump is aborted.
sendbackup: error [/sbin/dump returned 1]
I hope anyone can help me,
regards,
Pd. sorry my english...
Luis
--
Frank Smith [EMAIL PROTECTED]
Systems Administrator Voice: 512-374-4673
Hoover's Online Fax: 512-374-4501
Encryption of data
Title: Encryption of data Is it possible to encrypt the data written to tape with Amanda? I'm thinking of PGP or GPG so that off site storage security breaches are not as much of a concern. Thanks, Trevor.
amdump problem
I have a problem with my server, i try configure amanda, buy I get this mail
when run amdump:
These dumps were to tape hosting01.
The next tape Amanda expects to use is: a new tape.
FAILURE AND STRANGE DUMP SUMMARY:
ns.webspac /usr/local/backup/data lev 0 FAILED [/sbin/dump returned 1]
STATISTICS:
Total Full Daily
Estimate Time (hrs:min)0:00
Run Time (hrs:min) 0:01
Dump Time (hrs:min)0:00 0:00 0:00
Output Size (meg) 0.00.00.0
Original Size (meg) 0.00.00.0
Avg Compressed Size (%) -- -- --
Filesystems Dumped0 0 0
Avg Dump Rate (k/s) -- -- --
Tape Time (hrs:min)0:00 0:00 0:00
Tape Size (meg) 0.00.00.0
Tape Used (%) 0.00.00.0
Filesystems Taped 0 0 0
Avg Tp Write Rate (k/s) -- -- --
FAILED AND STRANGE DUMP DETAILS:
/-- ns.webspac /usr/local/backup/data lev 0 FAILED [/sbin/dump returned 1]
sendbackup: start [ns.xxx.com:/usr/local/backup/data level 0]
sendbackup: info BACKUP=/sbin/dump
sendbackup: info RECOVER_CMD=/sbin/restore -f... -
sendbackup: info end
| DUMP: You can't update the dumpdates file when dumping a subdirectory
| DUMP: The ENTIRE dump is aborted.
sendbackup: error [/sbin/dump returned 1]
I hope anyone can help me,
regards,
Pd. sorry my english...
Luis
Re: First time Changer user
On Thursday 06 February 2003 13:40, Chris Dahn wrote: >> I had and was watching this comedy >> at a remote terminal with the st.conf changes already entered in >> my vi session. When asked to assist, I saved my changes and >> quit, touched /reconfigure and logged off. Then I walked over to >> the console where they were working and started a reboot. I >> walked away before the reboot completed saying it "will work >> now". >> >> arrogant bastard jon > > I applaud you. In fact, I think we should nominate Jon for this > year's BOFH. Chuckle... I was gonna make some remark about how one should always act when in front of the "frogs", but I like this one better. Its choice. Any sufficiently advanced technology is indestinguishable from magic, to borrow from Asimov. Great one Jon, keep up the good work. -- Cheers, Gene AMD K6-III@500mhz 320M Athlon1600XP@1400mhz 512M 99.23% setiathome rank, not too shabby for a WV hillbilly
dds4 drive
Hello, Any suggestions on where to buy a dds4 drive at a good price? Soo
Ultrium-LTO tapetype
Has anybody setup an Exabyte LTO tape drive with amanda? If so how do
you determine the "filemark" and "speed" entries in the tapetyep
section.
Here is what I have so far:
define tapetype Ultrium-LTO {
comment "Exabyte LTO 100GB/200GB tape drive"
length 204800 mbytes
filemark ?
speed?
}
Thanks for any information.
-Ben
Re: First time Changer user
> I had and was watching this comedy > at a remote terminal with the st.conf changes already entered in my vi > session. When asked to assist, I saved my changes and quit, touched > /reconfigure and logged off. Then I walked over to the console where they > were working and started a reboot. I walked away before the reboot > completed saying it "will work now". > > arrogant bastard jon I applaud you. In fact, I think we should nominate Jon for this year's BOFH. -- |Christopher Dahn | Software Engineering Research Group,| |3141 Chestnut St. | Feel the SERG!| |Attn: CS Dept., CAT 186 | http://serg.cs.drexel.edu/ | |Philadephia, PA 19104 -- -- --
Re: First time Changer user
On Thu, 6 Feb 2003 at 5:25pm, David Flood wrote First thing's first -- could we trim a bit here folks. Furrfu. > Firstly thankyou Jon and Gene for answering my first email. Sorry I forgot to >specify my > setup. It's as follows: > > amanda 2.4.3 > Solaris 8 > Overland DLT80 (10 tape changer) > > Under /dev/rmt I only have '0' which is the one I'm using to get access to the >drive. > What is mtx and is it possible I could use this? There was no examples of this file >under mtx is a bit of software used to control changers. See here: http://mtx.badtux.net/ > the examples directory of the amanda distribution. I am looking at sgen and thus far mtx is used with the chg-zd-mtx changer described in docs/TAPE.CHANGERS and documented in the changer script itself, which gets installed into /usr/local/libexec by default. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
RE: First time Changer user
h ithere, had your same problem with a Sun Storedge L20 tape loader with 2 drives: heres your solution: run contrib/config_sgen_solaris.sh You'll be using the sgen(7D) generic SCSI device driver for the robot, and end up with a device like /dev/scsi/changer/cXtYd0. enjoy liam -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Jon LaBadie Sent: 06 February 2003 18:56 To: [EMAIL PROTECTED] Subject: Re: First time Changer user On Thu, Feb 06, 2003 at 05:25:02PM -, David Flood wrote: > On 5 Feb 2003 at 14:20, Jon LaBadie wrote: > > Under /dev/rmt I only have '0' which is the one I'm using to get access to the drive. > What is mtx and is it possible I could use this? There was no examples of this file under mt (Magnetic Tape) is supplied with solaris and manipulates the tape drive itself mtx (Magnetic Tape eXchanger :) is freeware and manipulates tape changers. See freshmeat.net and/or sourceforge.net > the examples directory of the amanda distribution. I am looking at sgen and thus far > don't really understand it but will keep trying. The method I used for setting up my > autochanger was the one I thought was used for any new device attached to a sun: > > shutdown > add device > take system up to ok promt > boot -r Not, if as Gene mentioned, the system does not scan the appropriate id/lun. As I noted, read the drive installation very carefully and > > In my case it even involved getting to the drive itself (removing the case) > > and setting some switches not accessible from outside. It also required mods > > to the "st" (scsi tape) driver configuration file, /kernel/drv/st.conf. Your installation manual should provide similar directions. The one Sun dlt drive I installed for a client (not a changer) had very detailed directions regarding changes needed to st.conf. I was asked to look at it because the technicians sent by the Sun Reseller could not figure out how to get it working. They were sure something was broken. Of course they had been too smart to Read The Fine Manual. I had and was watching this comedy at a remote terminal with the st.conf changes already entered in my vi session. When asked to assist, I saved my changes and quit, touched /reconfigure and logged off. Then I walked over to the console where they were working and started a reboot. I walked away before the reboot completed saying it "will work now". arrogant bastard jon -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: question
--On Thursday, February 06, 2003 12:48:44 -0500 [EMAIL PROTECTED] wrote: My tape server only has a tape driver with a single slot. We have to manually change tapes when do backup. So far, amanda doesn't ask user to change tape when a tape is full. The 'amdump' process just exits and write error message to log. Is this a configuration issue? What should I do? Option 1: Make 'runtapes' greater than 1 in your config. Configure a changer (use 'chg-manual' to manually change tapes). Note: You have to be there to change the tape, the dumps stop until you do. Option 2: Configure a large holdingdisk. Set 'reserve' to a low number (or 0) Backups that don't fit on your tape will keep going to the holdingdisk, then you run amflush to write them out to tape (if that tape fills, you can keep running amflush until it all ends up on a tape). Frank -- Frank Smith [EMAIL PROTECTED] Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501
Re: question
On Thu, 6 Feb 2003 at 12:26pm, [EMAIL PROTECTED] wrote > My tape server has a tape driver with a single slot. > I use the tape driver to backup several client hosts. > I have two questions after tests: > 1. Why 'amdump' doesn't ask to change tape when a tape is full? Amanda can't span tapes with a single DLE. If you want to use more than one tape per amdump, you'll need to define a changer. With your setup, you can use chg-manual, and amdump will prompt you for a new tape. You must also set runtapes > 1. > 2. How let 'amdump' to override the previous backup? > For example, when I ran 'amadmin DailySet1 find cat /boot', I saw > multiple > "cat /boot" in VOL0001. I really want the latest. I don't understand what you mean by override. If you have multiple entries for a single DLE on one tape (I think that's what your example is trying to say), then that must be the result of an amflush. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: First time Changer user
On Thu, Feb 06, 2003 at 05:25:02PM -, David Flood wrote: > On 5 Feb 2003 at 14:20, Jon LaBadie wrote: > > Under /dev/rmt I only have '0' which is the one I'm using to get access to the >drive. > What is mtx and is it possible I could use this? There was no examples of this file >under mt (Magnetic Tape) is supplied with solaris and manipulates the tape drive itself mtx (Magnetic Tape eXchanger :) is freeware and manipulates tape changers. See freshmeat.net and/or sourceforge.net > the examples directory of the amanda distribution. I am looking at sgen and thus far > don't really understand it but will keep trying. The method I used for setting up my > autochanger was the one I thought was used for any new device attached to a sun: > > shutdown > add device > take system up to ok promt > boot -r Not, if as Gene mentioned, the system does not scan the appropriate id/lun. As I noted, read the drive installation very carefully and > > In my case it even involved getting to the drive itself (removing the case) > > and setting some switches not accessible from outside. It also required mods > > to the "st" (scsi tape) driver configuration file, /kernel/drv/st.conf. Your installation manual should provide similar directions. The one Sun dlt drive I installed for a client (not a changer) had very detailed directions regarding changes needed to st.conf. I was asked to look at it because the technicians sent by the Sun Reseller could not figure out how to get it working. They were sure something was broken. Of course they had been too smart to Read The Fine Manual. I had and was watching this comedy at a remote terminal with the st.conf changes already entered in my vi session. When asked to assist, I saved my changes and quit, touched /reconfigure and logged off. Then I walked over to the console where they were working and started a reboot. I walked away before the reboot completed saying it "will work now". arrogant bastard jon -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
question
My tape server only has a tape driver with a single slot. We have to manually change tapes when do backup. So far, amanda doesn't ask user to change tape when a tape is full. The 'amdump' process just exits and write error message to log. Is this a configuration issue? What should I do? Thanks! David
Re: First time Changer user
On 5 Feb 2003 at 14:20, Jon LaBadie wrote: > On Wed, Feb 05, 2003 at 06:06:33PM -, David Flood wrote: > > We have had a tape changer running with amanda for a while but it was just being >used > > as the a single drive and the data is starting to exceed the size of the >tape.(DLT40 - > > 80G compressed) > > > > I'm starting to look into utilizing the functionality of the changer. I've looked >at the docs > > and have the impression that all I have to do is specify the chg-scsi in the >amanda.conf > > and set-up the chg-scsi.conf. Here are examples of the part of amanda.conf and > > chg-scsi.conf. > > > > amanda.conf: > > tpchanger "chg-scsi" # the tape-changer glue script > > tapedev "0" # the no-rewind tape device to be used > > changerfile "/opt/amanda/etc/amanda/homes/chg-scsi.conf" > > > > chg-scsi.conf: > > number_configs 1 > > eject 0 > > sleep 60 > > cleanmax 10 > > changerdev/dev/rsst4 > > # > > # Next comes the data for drive 0 > > # > > config0 > > drivenum 0 > > dev /dev/rmt/0bn > > startuse 0 > > enduse4 > > statfile /opt/amanda/etc/amanda/homes/tape0-slot > > cleancart 5 > > cleanfile /opt/amanda/etc/amanda/homes/tape0-clean > > agecount/opt/amanda/etc/amanda/homes/totaltimef > > tapestatus /opt/amanda/etc/amanda/homes/tape0-status > > > > > > I think I understand the contents of this file but I don't know what to put for >the > > changerdev file. Up till now I've been using /dev/rmt/0 for backing up to a single >tape so > > how do I find what this value should be for my changer.(/dev/rsst4 does not exist) > > > > Also when I've got the changer working how do the tape numbers work? If a night's > > backup tape starts on homes02 and spreads 2 tapes will amanda understand it's > > spread over homes02 and homes03? If so how does this affect restores? > > David, > gotta learn to mention amanda version and OS type. They may affect the replies. > > Seems to be Solaris you are using (a /opt and /dev/rmt). > > My changer, when installed according to HP's very good and specific directions, > resulted in two separate /dev/rmt devices, 0 for the tape drive, 1 for the changer > mechanism. When earlier, I installed it following "Jon's" instructions and not > RTFM'ing, I got one device. > > That doesn't mean you will always get 2 /dev/rmt devices, just that is one way > of working and urging you to get to the docs for your brand and model and OS. > In my case it even involved getting to the drive itself (removing the case) > and setting some switches not accessible from outside. It also required mods > to the "st" (scsi tape) driver configuration file, /kernel/drv/st.conf. > > If your solaris is a recent update of Solaris 8, Sun has tried to generalize > tape changer drivers. It uses something called the "sgen" driver. When > configured, changers will be sensed and devices created in /dev/???/changer/XXX. > I've not done the configuration because "don't fix what ain't broken". But I > think you can find stuff about setting it up on "docs.sun.com", man sgen, and > even this mail list archives on yahoo. > > Myself, I use mtx to manipulate the changer just like mt manipulates the drive. > And I use chg-mtx, not even chg-zd-mtx. It is simpler and it works for me. > > > As to your final questions, you will also have to set the amanda.conf parameter > "runtapes" to be more than 1. Amanda will use extra tapes only if necessary, > It will continue to use them in the same sequence it currently does, not skipping > any tapes because it reports "next tapes are 5 and 6" but it only uses 5. However > you may want to increase the number of tapes in rotation so you can still get > the same number of dumpcycles in the number of tapes. Note, you can have more > tapes in rotation than tapecycle specifies. That is a minimum. > > Recovery? No change. Amanda knows what it put on each tape. It asks you for > what tapes are needed. > Firstly thankyou Jon and Gene for answering my first email. Sorry I forgot to specify my setup. It's as follows: amanda 2.4.3 Solaris 8 Overland DLT80 (10 tape changer) Under /dev/rmt I only have '0' which is the one I'm using to get access to the drive. What is mtx and is it possible I could use this? There was no examples of this file under the examples directory of the amanda distribution. I am looking at sgen and thus far don't really understand it but will keep trying. The method I used for setting up my autochanger was the one I thought was used for any new device attached to a sun: shutdown add device take system up to ok promt boot -r - David Flood Systems Administrator [EMAIL PROTECTED] Tel: +44 (0)1224 262774 The Robert Gordon University School of Computing St. Andrews Street Aberdeen -
question
My tape server has a tape driver with a single slot. I use the tape driver to backup several client hosts. I have two questions after tests: 1. Why 'amdump' doesn't ask to change tape when a tape is full? 2. How let 'amdump' to override the previous backup? For example, when I ran 'amadmin DailySet1 find cat /boot', I saw multiple "cat /boot" in VOL0001. I really want the latest. Thanks advanced! David
Installing 2.4.4b1 on Cygwin
Hi! I'm trying to install 2.4.4b1 on a Win2K/Cygwin machine. I've run into a few questions, which may be blindingly obvious, but I'm totally missing... 1) which user/group: should I create a new user? There isn't a /dev directory, so I can't determine which group to put the user in, although since I'll be using tar, I'm not sure that's really important. So is it ok to just use --with-user=Administrator --with-group=SYSTEM? Do I need to change this to --with-user=W0083366\Administrator if I'm trying to specify a local admin vs a domain admin? 2) are there any odd things I need to do to set up xinetd to work? I haven't really played around with Cygwin that much -- is there a FAQ I could go to that will describe how to get xinetd up and running? Thanks muchly, Ricky Morse
Tapetype for Seagate DDS-4 drive
Does anyone know the tapetype for a Seagate STD6401 LW-RY DDS-4 external SCSI tape drive? Thanks for the help!!! __ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com
Re: Question about uninstall and reinstall
On Thu, Feb 06, 2003 at 05:21:59AM -0800, Pro-Fit wrote: > > I'm concerned that there might be some issues on one > of my clients > that had the whole amanda package (and an earlier > version at that) > installed as part of the Linux OS install. > > Here's what I'm thinking > > Machine A has new amanda and is configured to be the > backup server > (it has the tape drive). Machine B is the client. It > also has the > amanda software, but it was installed at an earlier > time, and was > configured to use a different --with-config, and it They need not be the same. > was setup to be a server. That is "mostly" which components get built/installed. Not functionallity. > > 1. Does the client thinks it's the server, even though > I can connect > to it from the main amanda server using amcheck? See above. -- Waiting for corrections if I'm in error. > 2. Do I need to completely remove amanda on machine b, > and run setup > using the without-server option along with the > with-index-server and > with-tape-server options? Check the docs/INSTALL file. There is very little needed on the client. And it was probably installed with the client parts also. I think the only significant thing from the server install is a couple of entries in inet/xinet.conf. Only one seems to be required for the client. Delete the server specific entries. > 3. Can I modify the client's amanda.conf file to point > to the main amanda server? Don't recall the client needing the config files. Corrections?? I don't seem to have a client about at the moment :(( > 4. If I need to uninstall amanda, is just a matter of > deleting all > instances of amanda (files and directories)? You mention linux. If installed from an rpm, that is an rpm issue, not amanda. Follow those procedures. If built from sources, there is an "uninstall" target in the makefile. But I think your idea is correct if you add to it the network config stuff (inet.conf and services files) and the crontab stuff and any password/group file stuff, and any /etc/aman... files like amandates. Any others? > 5. Can I run the latest version of amanda, and > *assume* it will overwrite the existing config? > The with-config that's on the client is the default > config. I've modified the with-config on the server install. If the install locations are the same and the user id are the same, I would. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Question about uninstall and reinstall
On Thu, 6 Feb 2003 at 5:21am, Pro-Fit wrote > Machine A has new amanda and is configured to be the > backup server > (it has the tape drive). Machine B is the client. It > also has the > amanda software, but it was installed at an earlier > time, and was > configured to use a different --with-config, and it > was setup to be > a server. > > 1. Does the client thinks it's the server, even though > I can connect > to it from the main amanda server using amcheck? Nope. > 2. Do I need to completely remove amanda on machine b, > and run setup > using the without-server option along with the > with-index-server and > with-tape-server options? No, you don't *have* to, but it can make things easier. > 3. Can I modify the client's amanda.conf file to point > to the main > amanda server? Clients don't ever touch an amanda.conf. You can just remove it. > 4. If I need to uninstall amanda, is just a matter of > deleting all > instances of amanda (files and directories)? That depends on how it was installed. If it was installed via something like rpm or apt, use those to uninstall it. > 5. Can I run the latest version of amanda, and > *assume* it will > overwrite the existing config? The with-config that's > on the client > is the default config. I've modified the with-config > on the server > install. Not necessarily, unless you know the ./configure options used to build the client version. I'd uninstall it if you plan on installing a new version. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: strange dump message
On Wed, 2003-02-05 at 11:07, Jon LaBadie wrote: > What I'm getting at, is if my guess is correct, 41021 is an inode num, > maybe you can find the offending file/directory and investigate. Two > commands to assist: the ls command has a "-i" option that prints the > inode number of the entries. Alternatively, if you don't have an idea > where the offender is, find can locate by inode number: > >find -xdev -inum 41021 > > The "-xdev" keeps the search to the one fs since each fs will have its > own inode 41021. > > The above syntax is for system V type commands. I presume linux with > its gnu counterparts has the same/similar syntax. John, Yep, that same syntax works here, and it turns out that inode 41021 is the directory entry /etc/mysql. Now this is why it's puzzling to me. This directory is where the Debian mysql-common package puts the sample my.cnf file. Nothing is actively being changed anywhere in this directory when a dump occurs, which was a possibility brought up by Martin Schwarz in a subsequent reply. Would temporarily renaming the directory cause its inode to change? How about copying it to a different file, deleting the original, then renaming the copy to the original name? I'm trying to think of something I could do that would not require a reboot/fsck, but would get the directory off of that inode. Thanks, Eric
Question about uninstall and reinstall
Hello all, another amanda newbie here (oh geesh, I know...) I've got it setup, just some questions about the client end. Any thoughts would be greatly appreciated! I'm concerned that there might be some issues on one of my clients that had the whole amanda package (and an earlier version at that) installed as part of the Linux OS install. Here's what I'm thinking Machine A has new amanda and is configured to be the backup server (it has the tape drive). Machine B is the client. It also has the amanda software, but it was installed at an earlier time, and was configured to use a different --with-config, and it was setup to be a server. 1. Does the client thinks it's the server, even though I can connect to it from the main amanda server using amcheck? 2. Do I need to completely remove amanda on machine b, and run setup using the without-server option along with the with-index-server and with-tape-server options? 3. Can I modify the client's amanda.conf file to point to the main amanda server? 4. If I need to uninstall amanda, is just a matter of deleting all instances of amanda (files and directories)? 5. Can I run the latest version of amanda, and *assume* it will overwrite the existing config? The with-config that's on the client is the default config. I've modified the with-config on the server install. Thank you all very much!!! Michael __ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com
Re: Running AMANDA over the Internet
There is support for kerberos, which does two things authenticates the request from the server to the client encrypts the backup stream The first step is actually quite important. Without it, someone else can run a backup and get your bits sent to them, even if they are protected from eavesdropping while being sent to the attacker. Kerberos support is currently not in good shape. You have to get the krb bits separately (separate CVS module), and they aren't tagged consistently with releases. I have tried to compile them recently, and had trouble. As others have said, using Kerberos requires, in addition to getting amanda+krb to build correctly, having a Kerberos server and the clue to administer the realm. Using IPsec seems like a good method, since it can provide authentication as well. [The following assumes famililiarity with RFC2401 and discusses BSD systems.] While others have mentioned VPN, one can instead transport mode ESP. The trick is getting Security Policy Database (SPD) entries to match the amanda traffic. Since ports change for the backup stream, it might be best to set per-socket policy in the amanda programs using setsockopt(IP_IPSEC_POLICY). This approach is likely workable on Solaris. Last I checked, the FreeSWAN IPsec implementation for Linux did not support an SPD as requires by RFC2401, and in particular using ports as selectors did not work. One can also just use tunnel-mode ESP for all traffic to the particular remote host. Gregory Troxel <[EMAIL PROTECTED]>
Re: Amflush failing silently
Hi, Thanks, but that can't be the problem, because this patch had been applied *before* the last failing amflush. And it does not show the described symptoms, amflush is not hanging and not eating up memory and cpu-cycles, it simply terminates fast without sending an e-mail or doing anything it is expected to do. Noe core-files laying around, it just marks the first matching tape as used in the database and then terminates. Any thoughts? Christoph Jean-Louis Martineau schrieb: > On Wed, Feb 05, 2003 at 10:14:35AM +0100, Christoph Scheeder wrote: > >>Hi, >>i have a problem with an amanda-2.4.3 installation and amflush. >>Problem >>multiple amanda-runs went to disk cause of tape-errors. >>only last run out of 3 shall be flushed to tape. >>Problem if i tell amflush only to flush run named "C:", for example, >>amflush does nothing, terminating without sending an e-mail. >>after that i have to run amcleanup, and in the generated e-mail it tells me >>there where no files to flush, but the tape in the drive is marked as used. >>If i move the other dir's out of the holding-area and do an amflush >>all works fine. >> >>AFAIR a similar problem was reported a few month back, but i couldn't find >>the messages in the mailinglist archives. > > > Read the 'double frees' thread on the amanda-hackers mailing list. > > Jean-Louis
