On Tuesday 07 November 2006 10:38, Thomas Traeger wrote: > Hello list, > > I posted this before but as I was no member of bacula-users it did not > (yet) go through the moderator filter. > > We are using bacula for around 9 months now and are so far very happy > with it. 2 weeks ago we installed bacula on a new server, a Fujitsu > Siemens RX300R3, the jukebox is still a Dell PowerVault 122T with a > Quantum SDLT320 drive. We experienced a occasional server crash during > backup 3 times now :o(. The last message found in bconsole after a reset > is always something like this: > > 03-Nov 00:43 pdc02-sd: backup_ora01.2006-11-02_21.15.02 Error: > block.c:538 Write error at 303:4210 on device "Quantum_SDLT320" > (/dev/nst0). ERR=Device or resource busy. > 03-Nov 00:43 pdc02-sd: backup_ora01.2006-11-02_21.15.02 Error: Re-read > of last block OK, but block numbers differ. Last block=166408 Current > block=0. > 03-Nov 00:43 pdc02-sd: End of medium on Volume "IFS_daily_2" > Bytes=303,252,171,056 Blocks=4,700,710 at 03-Nov-2006 00:43. > > 07-Nov 02:01 pdc02-sd: backup_pdc01.2006-11-06_23.59.00 Error: > block.c:538 Write error at 6:8639 on device "Quantum_SDLT320" > (/dev/nst0). ERR=Device or resource busy. > 07-Nov 02:01 pdc02-sd: Re-read of last block succeeded. > 07-Nov 02:01 pdc02-sd: End of medium on Volume "Server_daily_1" > Bytes=6,556,780,347 Blocks=101,639 at 07-Nov-2006 02:01. > > Once I found the following logentry in /var/log/messages: > > Nov 3 00:43:16 ifs01 kernel: Unable to handle kernel paging request at > virtual address 0524c3f0 > > After the first crash of this kind the even Dell PowerVault stopped > working and had to be replaced... > > We are using SLES 10 with the latest patches, bacula 1.38.11-3 installed > using the rpm packages provided for Suse 10.1. > > Is anyone else experiencing such strange things? Is there a connection > to recently reported problems with a stock Suse 10.1 Kernel? AFAIK SLES > 10 is based on Suse 10.1. >
As Arno pointed out this is probably a SuSE kernel problem. From the output you have posted above, it looks identical to the problem that I reported to them. I resolved the problem here by upgrading to their 10.2 kernel. My reading of the situation based on incomplete data is: 1. This is a SuSE specific problem. 2. They don't seem to be treating this as an important problem. I suggest you use all means possible to complain to SuSE and Novell about this problem. Specifically, you can start by expressing your concern in the bug report. I find it totally unacceptable that a so called "serious" Linux OS provider would allow such a low level, critcal bug to exist in their system for more than a couple of days after being informed of it. Bug reported 28 September, still unresolved. During testing of this bug (fortunately on a test system rather than my developement system), the bug totally trashed my hard disks (total loss). https://bugzilla.novell.com/show_bug.cgi?id=208782 Best regards, Kern ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
