RE: help on DAT 40
http://mtx.sourceforge.net 'tapeinfo' -> poll all the settings from userspace->kernelspace->firmware&eeprom 'scsitape' -> set various bits and pieces of above The actuall 'mtx' program is a must if you have a tape changer as well. -dan > -Original Message- > From: Maurizio Bossi [mailto:[EMAIL PROTECTED]] > Sent: Friday, March 15, 2002 10:59 AM > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED] > Subject: help on DAT 40 > > > Hi .. > I have the same problem of compression with HP c5687 DAT40 > and Linux, could > you send me a copy of the debugging software of Eric Lee > Green to check the > device ? > I used "mt" but I don't have much information > Do you have information on dip-switch more than in manual for > compression ? > > Regards >
RE: DAT40 problem
> -Original Message- > From: Maurizio Bossi [mailto:[EMAIL PROTECTED]] > Sent: Thursday, March 14, 2002 10:43 AM > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED] > Subject: DAT40 problem > > > Hi .. > I have the same problem of compression with HP c5687 DAT40 > and Linux, could > you send me a copy of the debugging software of Eric Lee > Green to check the > device ? > I used "mt" but I don't have much information > Do you have information on dip-switch more than in manual for > compression ? > > Regards http://mtx.sourceforge.net/ 'tapeinfo' MANPAGE: http://linuxcommand.org/man_pages/tapeinfo1.html Dumps out all the settings from userland, through the kernel, to the firmware of the device. 'scsitape' MANPAGE: http://linuxcommand.org/man_pages/scsitape1.html Configures many of the above parameters, including the default compression mode. hp.com has .pdfs of all thier manuals. -Dan Garthwaite
RE: Tapetype: STD2401LW
Those are some good speeds. What scsi card are you using, and what scsi
devices are on them? What's board/chip combo is the machine?
They aren't suspiciously fast, but I haven't been able to replicate them.
BTW- I find the email list atrocious, and consider it a read only device.
-dan
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> [EMAIL PROTECTED]
> Sent: Thursday, March 07, 2002 12:49 PM
> Subject: Re: Tapetype: STD2401LW
>
>
> Thanks to all nine of you (boy do I feel dumb) who pointed out that
> hardware compression should have been off. I forgot that
> rebooting this
> machine reset the drive to hardware compression on, argh.
> Five minutes
> with a screwdriver and no more... :)
>
> Some more realistic results:
>
>
> define tapetype unknown-tapetype {
> comment "Seagate STD2401LW DSS4 20/40GB DAT Drive"
> length 19457 mbytes
> filemark 1 kbytes
> speed 2681 kps
> }
>
> One other note... for two days now I've been getting timeouts
> on the primary MX server for the amanda list (
> mx-1.omniscient.com ). Has anyone else seen these issues?
> I've been able to send mail by connecting via telnet to port
> 25 and entering my headers and data block by hand, but I'd
> prefer not to have to do that ;) Is is er... Is it just
> my end, and some flaky MTA, oir is mx-1
> (surly.omniscient.com) under a heavy load so much of the time
> that the MTA times out? I noticed telnetting in that it took
> 15 seconds to get a connection... rather long,. but who's to
> say it's not some network issues on my end...
>
> Thanks again to everyone for pointing out the error of my way :)
>
> --
> Mark Hazen DataBuilt, Inc. (843) 836-2101 Ext. 251
>
> They may forget what you said, but they will never forget
> how you made them feel. --Carl W. Buechner
>
>
RE: Amanda and autoloader
No real problem. The way amanda talks to autoloaders is so simplistic, that it's easy to roll your own interface script. Not that I had to, three of the provided scripts that come with amanda all worked with my HP 1/20 (with some TLC). Make sure to install mt-st ( provides /usr/bin/mtx ). I think I had to tweak the version of amanda that RedHat ships with during testing, but I would recommend grabbing the newest stable source-tarball if your going to backup any non-RH unix machines, and compile a client from that tarball on every target architecture. IMHO: Amanda comes with too many 'gotchas', but is more than worth them once your past them. I'm backing up 11 HPUX machines, 8 RedHat machines, and three floors of PCs. -Dan > -Original Message- > From: Mats Blomstrand [mailto:[EMAIL PROTECTED]] > Sent: Tuesday, February 26, 2002 3:14 AM > To: [EMAIL PROTECTED] > Subject: Amanda and autoloader > > > Hi Daniel > > I saw your name on the amanda-users list. I would like to ask > a question. > > Im about to buy machines for backup. I have found that I can bring > the cost down quite a bit if I can use Amanda. Im a bit worried about > if I can manage to get amanda talking to the autoloader on a > new machine. > > Should I be worried? Is it a safe bet to get it working on a > HP 1/9, HP 1/20 or Dell 128T ? > > Please share any experience you might have. Thank you. > //Mats >
RE: amanda fails after upgrade
Sounds like a firewall to me: /etc/init.d/ipchains stop /etc/init.d/iptables stop -dan > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED]]On Behalf Of Joshua Baker-LePain > Sent: Thursday, January 24, 2002 1:30 PM > To: Ted Sariyski > Cc: [EMAIL PROTECTED] > Subject: Re: amanda fails after upgrade > > > On Thu, 24 Jan 2002 at 11:03am, Ted Sariyski wrote > > > Joshua Baker-LePain wrote: > > > > >Did you remember to add the amanda user? > > > > Yes, > > > > Even I can rsh from the server to the client without a password. > > > And is the UID the same as it was before? > > Did you do a "upgrade" install, or did you wipe the > filesystems and start > over? > > (P.S. Please keep amanda-users on the cc list, so everyone > can help out.) > > -- > Joshua Baker-LePain > Department of Biomedical Engineering > Duke University > > >
2.4.3b2 chg-scsi not compiled
As in it was skipped. I haven't investigated deeply enough to see if it was a './configure' option I missed. cd'ing into changer-src and 'make chg-scsi' built it, but then 'make install' didn't install it, 'cp' worked. Just reporting in. Should this go to amanda-hackers? -Dan Garthwaite
RE: schedual line 10: syntax error - HELP
And in that header, look at the value of the reply-to. Case in point, this message came to me, and not the list. -dan > -Original Message- > From: Jon LaBadie [mailto:[EMAIL PROTECTED]] > Sent: Saturday, December 08, 2001 11:06 PM > To: Dan Garthwaite > Subject: Re: schedual line 10: syntax error - HELP > > > On Thu, Dec 06, 2001 at 09:51:07AM -0500, Dan Garthwaite wrote: > > Why doesn't this list set the REPLY-TO to amanda-users? > > > >>> End of included message <<< > > It does when I get them. That is how I'm replying. > Here is a chunk of the headers I got on this mail: > > > From [EMAIL PROTECTED] Fri Dec 7 15:13:27 2001 > Received: from guiness.omniscient.com > (guiness.omniscient.com [64.134.101.78]) > ... > Reply-To: <[EMAIL PROTECTED]> > From: "Dan Garthwaite" <[EMAIL PROTECTED]> > To: "'Jeff Pratt'" <[EMAIL PROTECTED]> > Cc: <[EMAIL PROTECTED]> > Subject: RE: schedual line 10: syntax error - HELP > Date: Thu, 6 Dec 2001 09:51:07 -0500 > Message-Id: <001101c17e65$7081dc80$[EMAIL PROTECTED]> > > > > -- > Jon H. LaBadie [EMAIL PROTECTED] > JG Computing [EMAIL PROTECTED] > 4455 Province Line Road(609) 252-0159 > Princeton, NJ 08540-4322 (609) 683-7220 (fax) >
RE: schedual line 10: syntax error - HELP
whoa, I sent the message below three messages ago... and 24 hours ago. Damn Outlook and it's creators... -dan > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED]]On Behalf Of Dan Garthwaite > Sent: Thursday, December 06, 2001 4:41 PM > To: [EMAIL PROTECTED] > Subject: RE: schedual line 10: syntax error - HELP > > > amanda.mtg.saic.com //gagnona/F$ 11661 0 1970:1:1:0:0:0 0 0 > 1818845508 > > Nope, my bad, that is the eighth field. It has no right > being there at > all especially without the other degraded fields. > > Something has to be re-stated I think: all the smb clients > are allocated, > via samba lookups, before each run, and disklist is > dynamically created. > gagnona and isis-temp were in the old static disklist files before I > implemented the dynamic disklist, and thus exist in curinfo, > etc. Same goes > for isis-temp. > > So there is a real bug here, IMHO. Is anyone interested in > me attaching the > curinfo files? Let me know before amanda cycles through it > all ( 15 days ) > > I'm trying to catch up on the code and contribute. The more you read > driver.c and planner.c, the more impressive this system really is, and > remains relatively straightforward. Kudos all. I think, if > pushed, I'll > just check for out-of-bounds values in the 8th field in > driver.c, which is a > kludge and a band-aide. I'll put this off for now so I can > test some more. > > -dan > > > > >
RE: schedual line 10: syntax error - HELP
Why doesn't this list set the REPLY-TO to amanda-users? Great idea, I'll check. Your right that I'm not running them serially, but with a 5 minute windows out of cron. I'll write a wrapper around both to find out, and add some timestamping to my script's errlog. -dan > -Original Message- > From: Jeff Pratt [mailto:[EMAIL PROTECTED]] > Sent: Wednesday, December 05, 2001 6:51 PM > To: [EMAIL PROTECTED] > Subject: Re: schedual line 10: syntax error - HELP > > > Just a thought... are you sure your scripts are COMPLETING > before the > run??? > > Jeff > > Dan Garthwaite wrote: > > > Help! Short of strace'ing the entire amdump *shudder*, I > can't grep what > > the hell is causing this error. Any GUESSES even would be more than > > welcome. > >-dan > > > > > >>-Original Message- > >>From: [EMAIL PROTECTED] > >>[mailto:[EMAIL PROTECTED]]On Behalf Of Dan Garthwaite > >>Sent: Tuesday, December 04, 2001 10:46 AM > >>To: [EMAIL PROTECTED] > >>Subject: schedual line 10: syntax error > >> > >> > >>A moment of introduction: > >> I'm the UNIX administrator for a branch of Science Applications > >>International Corporation. I've been using amanda quite > >>happily now for ~6 > >>months to feed an HP SureStore Ultrium 1/20 ( that's 2TB > >>native ) backing up > >>a data center of RH7.1 and HPUX 10.20 boxen. > >> > >>The problem: > >> Extending this system now to back up ~100 windows lusers. > >>Half of these > >>are mobile laptops. To compensate for the on again, of again > >>nature of all > >>the laptops, I've written an nice awk/bash script to autogenerate > >>/etc/amandapass and append the lusers to > >>/etc/amanda/Daily/disklist on the > >>fly, just before amcheck and amdump run. > >> > >> I'm pretty sure the scripts run flawlessly. One of the > >>features of the > >>scripts is to append a cylinder label ( 100, 101, 102... ), > >>the same to each > >>windows host. ( to keep the 16 dumpers from hitting the same > >>poor laptop ) > >>This also seems to work well during test runs. > >> > >> The problem, adding the whole list of 60 or so clients, the > >>whole dump run > >>aborts. The logs in /tmp/amanda/*$(date +%Y%m%d).debug all > >>check out, look > >>fine, with no errors. The logs in /var/lib/Daily are > another matter: > >> > >> [amanda@amanda Daily]$ tail amdump.1 > >> process6.mtg.saic.com / 0 1 2001:11:20:22:15:7 271 18 > >> process6.mtg.saic.com /stand 0 1 2001:11:20:22:58:8 20 0 > >> morpheus.mtg.saic.com /stand 0 1 2001:11:20:22:15:37 14 0 > >> > >> driver: schedule line 10: syntax error > >> driver: adding holding disk 0 dir /var/lib/amanda/dumps > size 97819744 > >> reserving 97819744 out of 97819744 for degraded-mode dumps > >> taper: DONE [idle wait: 629.108 secs] > >> taper: writing end marker. [AL5853 OK kb 0 fm 0] > >> amdump: end at Mon Dec 3 17:15:42 EST 2001 > >> > >>and > >> [amanda@amanda Daily]$ tail log.20011203.0 > >> FAIL planner amanda.mtg.saic.com //wiggin_jo/C$ 0 [dumps > too big, but > >>cannot incremental dump new disk] > >> FAIL planner amanda.mtg.saic.com //win2klinux/C$ 0 [dumps > >>too big, but > >>cannot incremental dump new disk] > >> FAIL planner amanda.mtg.saic.com //jagoe_d/C$ 0 [dumps too > >>big, but cannot > >>incremental dump new disk] > >> FAIL planner amanda.mtg.saic.com //saic37/C$ 0 [dumps too > >>big, but cannot > >>incremental dump new disk] > >> FAIL planner amanda.mtg.saic.com //dan_ny3/G$ 0 [dumps too > >>big, but cannot > >>incremental dump new disk] > >> FAIL planner amanda.mtg.saic.com //esri-1/D$ 0 [dumps too > >>big, but cannot > >>incremental dump new disk] > >> FAIL planner amanda.mtg.saic.com //perkins_d/D$ 0 [dumps > too big, but > >>cannot incremental dump new disk] > >> FINISH planner date 20011203 > >> FATAL driver schedule line 10: syntax error > >> INFO taper tape AL5853 kb 0 fm 0 [OK] > >> > >>Where can I debug the 'schedual line 10: syntax error' in > >>both of these > >>files? > >> > >> -Dan Garthwaite > >> > >> > >> > > > > > > >
RE: schedual line 10: syntax error - HELP
In the section "GENERATING SCHEDULE:" in amdump's log the input to
read_schedule() in driver.c? If so, then I have the erroneous lines (
depending if 'line' the counter var is 0 or 1 based )
GENERATING SCHEDULE:
dozer.mtg.saic.com /data1 11661 0 1970:1:1:0:0:0 26459758 881991
amanda.mtg.saic.com //danny1/E$ 11661 0 1970:1:1:0:0:0 12179448 405981
amanda.mtg.saic.com //dan_ny3/C$ 11661 0 1970:1:1:0:0:0 8974904 299163
amanda.mtg.saic.com //isis_temp/E$ 11661 0 1970:1:1:0:0:0 6369703 212323
amanda.mtg.saic.com //saic94/H$ 11661 0 1970:1:1:0:0:0 6116400 203880
amanda.mtg.saic.com //tracey_greg/C$ 11661 0 1970:1:1:0:0:0 2797452 93248
amanda.mtg.saic.com //deb_kestler2/C$ 11661 0 1970:1:1:0:0:0 626388 20879
amanda.mtg.saic.com //allen_b/D$ 11661 0 1970:1:1:0:0:0 2358 78
amanda.mtg.saic.com //saic94/F$ 11661 0 1970:1:1:0:0:0 46 1
amanda.mtg.saic.com //gagnona/F$ 11661 0 1970:1:1:0:0:0 0 0 1818845508
amanda.mtg.saic.com //isis_temp/F$ 11661 0 1970:1:1:0:0:0 0 0 1818845508
cypher.mtg.saic.com /cypher_sd13 37 1 2001:10:15:1:42:13 182 10
cypher.mtg.saic.com /cypher_sd10 21 3 2001:11:30:1:0:25 292258 257
Notice the identical, and HUGE, number in the last field of gagnona and
isis_temp? This sets off flags in my book. The error, of course, is
'schedule line 10: syntax error', which coincides with this large,
repetitive number.
Those two clients are the only ones to show such a huge number in that field
as well. I'm on the scent, I know it. b)
AHA
skip_whitespace(s, ch); /* find the time number */
if(ch == '\0' || sscanf(s - 1, "%ld", &time) != 1) {
error("schedule line %d: syntax error", line);
continue;
}
skip_integer(s, ch);
Here is the field with the huge number. I guess these two hosts are timing
out.
I'm off to go see what the client's cifs/tcp settings are.
-dan
RE: schedual line 10: syntax error - HELP
amanda.mtg.saic.com //gagnona/F$ 11661 0 1970:1:1:0:0:0 0 0 1818845508 Nope, my bad, that is the eighth field. It has no right being there at all especially without the other degraded fields. Something has to be re-stated I think: all the smb clients are allocated, via samba lookups, before each run, and disklist is dynamically created. gagnona and isis-temp were in the old static disklist files before I implemented the dynamic disklist, and thus exist in curinfo, etc. Same goes for isis-temp. So there is a real bug here, IMHO. Is anyone interested in me attaching the curinfo files? Let me know before amanda cycles through it all ( 15 days ) I'm trying to catch up on the code and contribute. The more you read driver.c and planner.c, the more impressive this system really is, and remains relatively straightforward. Kudos all. I think, if pushed, I'll just check for out-of-bounds values in the 8th field in driver.c, which is a kludge and a band-aide. I'll put this off for now so I can test some more. -dan
RE: schedual line 10: syntax error - HELP
A little more light here: The F$ on both the problem clients are *drum roll* empty. With the exception of the RECYCLED hidden directory. Anyone ever have problems with planner backing up empty partitions? -dan
RE: Verifying backups
> > Right now we use a shell script to gather information about date, file > size, uid, gid and permission. We run teh script on the server and on > the restored replica. We filter both lists with a python script and > parses the through diff to see how much they differ. > > This is very time consuming as you have to restore a complete server > each time. I have another idea to a script which could be run > from cron: > There are a a number of IDS (intrusion detection systems) that gathers the same info, but at the inode level, already optimized for speed. I settled on 'integrit', after fiddling with 'aide' for a month or so. 'tripwire' isn't an option for this shop. integrit already has the ability to compare two logs, remember to filter out inode changes... Maybe an option for md5sum'ing each partition's tape stream to be stored in curinfo? -dan
RE: schedual line 10: syntax error - HELP
Of course. Joshua Baker-LePain sent some suggestions to the list on this as well. I'm searching through the archives now, without capturing the output to/from PLANNER the source code isn't revealing me anything. Oh, the script takes less than a minute to complete, and I'm giving it a five minute window. No time conflict there. -dan > -Original Message- > From: Jeff Pratt [mailto:[EMAIL PROTECTED]] > Sent: Thursday, December 06, 2001 12:23 PM > To: [EMAIL PROTECTED] > Subject: Re: schedual line 10: syntax error - HELP > > > For some reason I've become insanely interested in this. Could you > update me with what you find out? > > thanks > > Jeff > > Dan Garthwaite wrote: > > > Why doesn't this list set the REPLY-TO to amanda-users? > > > > Great idea, I'll check. Your right that I'm not running > them serially, but > > with a 5 minute windows out of cron. I'll write a wrapper > around both to > > find out, and add some timestamping to my script's errlog. > > > > -dan > > > > > >>-Original Message- > >>From: Jeff Pratt [mailto:[EMAIL PROTECTED]] > >>Sent: Wednesday, December 05, 2001 6:51 PM > >>To: [EMAIL PROTECTED] > >>Subject: Re: schedual line 10: syntax error - HELP > >> > >> > >>Just a thought... are you sure your scripts are COMPLETING > >>before the > >>run??? > >> > >>Jeff > >> > >>Dan Garthwaite wrote: > >> > >> > >>>Help! Short of strace'ing the entire amdump *shudder*, I > >>> > >>can't grep what > >> > >>>the hell is causing this error. Any GUESSES even would be > more than > >>>welcome. > >>> -dan > >>> > >>> > >>> > >>>>-Original Message- > >>>>From: [EMAIL PROTECTED] > >>>>[mailto:[EMAIL PROTECTED]]On Behalf Of Dan Garthwaite > >>>>Sent: Tuesday, December 04, 2001 10:46 AM > >>>>To: [EMAIL PROTECTED] > >>>>Subject: schedual line 10: syntax error > >>>> > >>>> > >>>>A moment of introduction: > >>>>I'm the UNIX administrator for a branch of Science Applications > >>>>International Corporation. I've been using amanda quite > >>>>happily now for ~6 > >>>>months to feed an HP SureStore Ultrium 1/20 ( that's 2TB > >>>>native ) backing up > >>>>a data center of RH7.1 and HPUX 10.20 boxen. > >>>> > >>>>The problem: > >>>>Extending this system now to back up ~100 windows lusers. > >>>>Half of these > >>>>are mobile laptops. To compensate for the on again, of again > >>>>nature of all > >>>>the laptops, I've written an nice awk/bash script to autogenerate > >>>>/etc/amandapass and append the lusers to > >>>>/etc/amanda/Daily/disklist on the > >>>>fly, just before amcheck and amdump run. > >>>> > >>>>I'm pretty sure the scripts run flawlessly. One of the > >>>>features of the > >>>>scripts is to append a cylinder label ( 100, 101, 102... ), > >>>>the same to each > >>>>windows host. ( to keep the 16 dumpers from hitting the same > >>>>poor laptop ) > >>>>This also seems to work well during test runs. > >>>> > >>>>The problem, adding the whole list of 60 or so clients, the > >>>>whole dump run > >>>>aborts. The logs in /tmp/amanda/*$(date +%Y%m%d).debug all > >>>>check out, look > >>>>fine, with no errors. The logs in /var/lib/Daily are > >>>> > >>another matter: > >> > >>>>[amanda@amanda Daily]$ tail amdump.1 > >>>>process6.mtg.saic.com / 0 1 2001:11:20:22:15:7 271 18 > >>>>process6.mtg.saic.com /stand 0 1 2001:11:20:22:58:8 20 0 > >>>>morpheus.mtg.saic.com /stand 0 1 2001:11:20:22:15:37 14 0 > >>>> > >>>>driver: schedule line 10: syntax error > >>>>driver: adding holding disk 0 dir /var/lib/amanda/dumps > >>>> > >>size 97819744 > >> > >>>>reserving 97819744 out of 97819744 for degraded-mode dumps > >>>>taper: DONE [idle wait: 629.108 secs] > >>>>taper: writing end marker. [AL5853 OK kb 0 fm 0] > >>>>amdump: end at Mon Dec 3 17:15:42 EST 2001 > >>>> > >>>>and > >>>>[amanda@amanda Daily]$ tail log.20011203.0 > >>>>FAIL planner amanda.mtg.saic.com //wiggin_jo/C$ 0 [dumps > >>>> > >>too big, but > >> > >>>>cannot incremental dump new disk] > >>>>FAIL planner amanda.mtg.saic.com //win2klinux/C$ 0 [dumps > >>>>too big, but > >>>>cannot incremental dump new disk] > >>>>FAIL planner amanda.mtg.saic.com //jagoe_d/C$ 0 [dumps too > >>>>big, but cannot > >>>>incremental dump new disk] > >>>>FAIL planner amanda.mtg.saic.com //saic37/C$ 0 [dumps too > >>>>big, but cannot > >>>>incremental dump new disk] > >>>>FAIL planner amanda.mtg.saic.com //dan_ny3/G$ 0 [dumps too > >>>>big, but cannot > >>>>incremental dump new disk] > >>>>FAIL planner amanda.mtg.saic.com //esri-1/D$ 0 [dumps too > >>>>big, but cannot > >>>>incremental dump new disk] > >>>>FAIL planner amanda.mtg.saic.com //perkins_d/D$ 0 [dumps > >>>> > >>too big, but > >> > >>>>cannot incremental dump new disk] > >>>>FINISH planner date 20011203 > >>>>FATAL driver schedule line 10: syntax error > >>>>INFO taper tape AL5853 kb 0 fm 0 [OK] > >>>> > >>>>Where can I debug the 'schedual line 10: syntax error' in > >>>>both of these > >>>>files? > >>>> > >>>> -Dan Garthwaite > >>>> > >>>> > >>>> > >>>> > >>> > >> > >> > > > > > > >
RE: schedual line 10: syntax error - HELP
Help! Short of strace'ing the entire amdump *shudder*, I can't grep what the hell is causing this error. Any GUESSES even would be more than welcome. -dan > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED]]On Behalf Of Dan Garthwaite > Sent: Tuesday, December 04, 2001 10:46 AM > To: [EMAIL PROTECTED] > Subject: schedual line 10: syntax error > > > A moment of introduction: > I'm the UNIX administrator for a branch of Science Applications > International Corporation. I've been using amanda quite > happily now for ~6 > months to feed an HP SureStore Ultrium 1/20 ( that's 2TB > native ) backing up > a data center of RH7.1 and HPUX 10.20 boxen. > > The problem: > Extending this system now to back up ~100 windows lusers. > Half of these > are mobile laptops. To compensate for the on again, of again > nature of all > the laptops, I've written an nice awk/bash script to autogenerate > /etc/amandapass and append the lusers to > /etc/amanda/Daily/disklist on the > fly, just before amcheck and amdump run. > > I'm pretty sure the scripts run flawlessly. One of the > features of the > scripts is to append a cylinder label ( 100, 101, 102... ), > the same to each > windows host. ( to keep the 16 dumpers from hitting the same > poor laptop ) > This also seems to work well during test runs. > > The problem, adding the whole list of 60 or so clients, the > whole dump run > aborts. The logs in /tmp/amanda/*$(date +%Y%m%d).debug all > check out, look > fine, with no errors. The logs in /var/lib/Daily are another matter: > > [amanda@amanda Daily]$ tail amdump.1 > process6.mtg.saic.com / 0 1 2001:11:20:22:15:7 271 18 > process6.mtg.saic.com /stand 0 1 2001:11:20:22:58:8 20 0 > morpheus.mtg.saic.com /stand 0 1 2001:11:20:22:15:37 14 0 > > driver: schedule line 10: syntax error > driver: adding holding disk 0 dir /var/lib/amanda/dumps size 97819744 > reserving 97819744 out of 97819744 for degraded-mode dumps > taper: DONE [idle wait: 629.108 secs] > taper: writing end marker. [AL5853 OK kb 0 fm 0] > amdump: end at Mon Dec 3 17:15:42 EST 2001 > > and > [amanda@amanda Daily]$ tail log.20011203.0 > FAIL planner amanda.mtg.saic.com //wiggin_jo/C$ 0 [dumps too big, but > cannot incremental dump new disk] > FAIL planner amanda.mtg.saic.com //win2klinux/C$ 0 [dumps > too big, but > cannot incremental dump new disk] > FAIL planner amanda.mtg.saic.com //jagoe_d/C$ 0 [dumps too > big, but cannot > incremental dump new disk] > FAIL planner amanda.mtg.saic.com //saic37/C$ 0 [dumps too > big, but cannot > incremental dump new disk] > FAIL planner amanda.mtg.saic.com //dan_ny3/G$ 0 [dumps too > big, but cannot > incremental dump new disk] > FAIL planner amanda.mtg.saic.com //esri-1/D$ 0 [dumps too > big, but cannot > incremental dump new disk] > FAIL planner amanda.mtg.saic.com //perkins_d/D$ 0 [dumps too big, but > cannot incremental dump new disk] > FINISH planner date 20011203 > FATAL driver schedule line 10: syntax error > INFO taper tape AL5853 kb 0 fm 0 [OK] > > Where can I debug the 'schedual line 10: syntax error' in > both of these > files? > > -Dan Garthwaite > >
schedual line 10: syntax error
A moment of introduction: I'm the UNIX administrator for a branch of Science Applications International Corporation. I've been using amanda quite happily now for ~6 months to feed an HP SureStore Ultrium 1/20 ( that's 2TB native ) backing up a data center of RH7.1 and HPUX 10.20 boxen. The problem: Extending this system now to back up ~100 windows lusers. Half of these are mobile laptops. To compensate for the on again, of again nature of all the laptops, I've written an nice awk/bash script to autogenerate /etc/amandapass and append the lusers to /etc/amanda/Daily/disklist on the fly, just before amcheck and amdump run. I'm pretty sure the scripts run flawlessly. One of the features of the scripts is to append a cylinder label ( 100, 101, 102... ), the same to each windows host. ( to keep the 16 dumpers from hitting the same poor laptop ) This also seems to work well during test runs. The problem, adding the whole list of 60 or so clients, the whole dump run aborts. The logs in /tmp/amanda/*$(date +%Y%m%d).debug all check out, look fine, with no errors. The logs in /var/lib/Daily are another matter: [amanda@amanda Daily]$ tail amdump.1 process6.mtg.saic.com / 0 1 2001:11:20:22:15:7 271 18 process6.mtg.saic.com /stand 0 1 2001:11:20:22:58:8 20 0 morpheus.mtg.saic.com /stand 0 1 2001:11:20:22:15:37 14 0 driver: schedule line 10: syntax error driver: adding holding disk 0 dir /var/lib/amanda/dumps size 97819744 reserving 97819744 out of 97819744 for degraded-mode dumps taper: DONE [idle wait: 629.108 secs] taper: writing end marker. [AL5853 OK kb 0 fm 0] amdump: end at Mon Dec 3 17:15:42 EST 2001 and [amanda@amanda Daily]$ tail log.20011203.0 FAIL planner amanda.mtg.saic.com //wiggin_jo/C$ 0 [dumps too big, but cannot incremental dump new disk] FAIL planner amanda.mtg.saic.com //win2klinux/C$ 0 [dumps too big, but cannot incremental dump new disk] FAIL planner amanda.mtg.saic.com //jagoe_d/C$ 0 [dumps too big, but cannot incremental dump new disk] FAIL planner amanda.mtg.saic.com //saic37/C$ 0 [dumps too big, but cannot incremental dump new disk] FAIL planner amanda.mtg.saic.com //dan_ny3/G$ 0 [dumps too big, but cannot incremental dump new disk] FAIL planner amanda.mtg.saic.com //esri-1/D$ 0 [dumps too big, but cannot incremental dump new disk] FAIL planner amanda.mtg.saic.com //perkins_d/D$ 0 [dumps too big, but cannot incremental dump new disk] FINISH planner date 20011203 FATAL driver schedule line 10: syntax error INFO taper tape AL5853 kb 0 fm 0 [OK] Where can I debug the 'schedual line 10: syntax error' in both of these files? -Dan Garthwaite
