From: David K. <xd...@li...> - 2003-07-20 10:17:25
|
Hi all, I am trying to get bacula to work on SGI Irix 6.5.20. I know that this platform is not supported yet, but I am not sure my problem is platform specific. I think it has nothing to do with the mentioned platform itself. I just compiled, installed, configured and started the daemons. From console started jobs end everytime with errors, claiming that the Volume of name "" could not be used. But I have no such volume defined. In the default pool was added only one volume with name of "hostname0001". Media had to be re-labeled to be able to start the job (relabelling takes about 6 hrs for me - is it normal ? Is there a way to see what label is on the tape instead of the one in the MySQL db ?) Tables "file" and "filename" have already tens of megabytes although it seems jobs were never completed. Even if I explicitly mount the only volume I get followin errors: pteryx-dir: NightlySave.2003-07-20_10.10.26 Error: catreq.c:348 Invalid Catalog request: CatReq Job=NightlySave.2003-07-20_10.10.26 GetVolInfo VolName= write=1 pteryx-sd: NightlySave.2003-07-20_10.10.26 Warning: mount.c:230 Wanted Volume "p teryx0001". Current Volume "" not acceptable because: 1990 Invalid Catalog Request: CatReq Job=NightlySave.2003-07-20_10.10.26 Get VolInfo VolName= write=1 pteryx-sd: Please mount Volume "pteryx0001" on Storage Device "SDT-10000" for Jo b NightlySave.2003-07-20_10.10.26 Use "mount" command to release Job. pteryx-dir: NightlySave.2003-07-20_10.10.26 Error: catreq.c:348 Invalid Catalog request: CatReq Job=NightlySave.2003-07-20_10.10.26 GetVolInfo VolName= write=1 pteryx-sd: NightlySave.2003-07-20_10.10.26 Warning: mount.c:230 Wanted Volume "p teryx0001". Current Volume "" not acceptable because: 1990 Invalid Catalog Request: CatReq Job=NightlySave.2003-07-20_10.10.26 Get VolInfo VolName= write=1 pteryx-sd: NightlySave.2003-07-20_10.10.26 Fatal error: Job canceled. Do you have some idea what I could do wrong ? Or, if you think that it is really a platform specific problem, do you have some constacts to the people porting Bacula to Irix ? Thanks in advance, David Komanek Charles Univeristy in Prague Faculty of Science CZ, Prague |
From: Kern S. <ke...@si...> - 2003-07-20 11:43:50
|
Hello, You didn't mention what version of Bacula you are running -- 1.30a I imagine. It seems to me that in that version, using spaces in the Volume name would produce errors like you are seeing. If this is your case, just avoid using spaces in Volume names. Use only letters and numbers as well as the hypen or underscore. Relabeling should take only a few minutes - the time to rewind the tape, write a label and re-read it. If you hung a totally blank tape and you have an older tape drive, it could take 6 hours because the tape will "run away" when Bacula tries to read it. Newer tape drives do not have this problem. To avoid it, write and EOF on the tape before trying to label it in Bacula using mt. I have built the Client program on Irix -- so it should work, but I have never built or tested the Director or Storage daemon on Irix. Are you using them on=20 Irix too? Best regards, Kern On Sun, 2003-07-20 at 12:17, David Komanek wrote: > Hi all, >=20 > I am trying to get bacula to work on SGI Irix 6.5.20. I know that this > platform is not supported yet, but I am not sure my problem is platform > specific. I think it has nothing to do with the mentioned platform itself= . >=20 > I just compiled, installed, configured and started the daemons. From > console started jobs end everytime with errors, claiming that the Volume > of name "" could not be used. But I have no such volume defined. In the > default pool was added only one volume with name of "hostname0001". Media > had to be re-labeled to be able to start the job (relabelling takes about > 6 hrs for me - is it normal ? Is there a way to see what label is on the > tape instead of the one in the MySQL db ?) Tables "file" and "filename" > have already tens of megabytes although it seems jobs were never > completed. >=20 > Even if I explicitly mount the only volume I get followin errors: >=20 >=20 > pteryx-dir: NightlySave.2003-07-20_10.10.26 Error: catreq.c:348 Invalid > Catalog > request: CatReq Job=3DNightlySave.2003-07-20_10.10.26 GetVolInfo VolName= =3D > write=3D1 > pteryx-sd: NightlySave.2003-07-20_10.10.26 Warning: mount.c:230 Wanted > Volume "p > teryx0001". > Current Volume "" not acceptable because: > 1990 Invalid Catalog Request: CatReq > Job=3DNightlySave.2003-07-20_10.10.26 Get > VolInfo VolName=3D write=3D1 > pteryx-sd: Please mount Volume "pteryx0001" on Storage Device "SDT-10000" > for Jo > b NightlySave.2003-07-20_10.10.26 > Use "mount" command to release Job. > pteryx-dir: NightlySave.2003-07-20_10.10.26 Error: catreq.c:348 Invalid > Catalog > request: CatReq Job=3DNightlySave.2003-07-20_10.10.26 GetVolInfo VolName= =3D > write=3D1 > pteryx-sd: NightlySave.2003-07-20_10.10.26 Warning: mount.c:230 Wanted > Volume "p > teryx0001". > Current Volume "" not acceptable because: > 1990 Invalid Catalog Request: CatReq > Job=3DNightlySave.2003-07-20_10.10.26 Get > VolInfo VolName=3D write=3D1 > pteryx-sd: NightlySave.2003-07-20_10.10.26 Fatal error: Job canceled. >=20 >=20 >=20 > Do you have some idea what I could do wrong ? Or, if you think that it is > really a platform specific problem, do you have some constacts to the > people porting Bacula to Irix ? >=20 > Thanks in advance, >=20 > David Komanek > Charles Univeristy in Prague > Faculty of Science > CZ, Prague >=20 >=20 > ------------------------------------------------------- > This SF.net email is sponsored by: VM Ware > With VMware you can run multiple operating systems on a single machine. > WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the > same time. Free trial click here: http://www.vmware.com/wl/offer/345/0 > _______________________________________________ > Bacula-devel mailing list > Bac...@li... > https://lists.sourceforge.net/lists/listinfo/bacula-devel |
From: David K. <xd...@li...> - 2003-07-20 17:09:16
|
Thank you very much for the quick response. Yes, I use 1.30a and I am trying to get it workong on Irix 6.5.20 as both client and server. My colleague uses it on SuSe Linux and recomended it to me as a great solution for backuping. I had some problems with compiling the source code but it seems it were only minor problems while Irix compiler I have does not support C99 extensions. It is possible I made some mistake in the code but the "reparations" seemed very trivial to me .... Well, back to the problem. I completely dropped and recreated all baculas tables in MySQL (4.0.13). Here is the console output: # ./bacula start Starting the Storage daemon Starting the File daemon Starting the Director daemon # console Connecting to Director pteryx:9101 1000 OK: pteryx-dir Version: 1.30a (10 May 2003) *setde Using default Catalog name=MyCatalog DB=bacula Enter new debug level: 5 Available daemons are: 1: Director 2: Storage 3: Client 4: All Select daemon type to set debug level (1-4): 4 Connecting to Storage daemon SDT-10000 at pteryx.natur.cuni.cz:9103 3000 OK setdebug=5 Connecting to Client pteryx-fd at pteryx:9102 2000 OK setdebug=5 *autodisplay on *list volumes Pool: Default +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRet ention | Recycle | Slot | +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ *list pools +--------+---------+---------+---------+----------+-------------+ | PoolId | Name | NumVols | MaxVols | PoolType | LabelFormat | +--------+---------+---------+---------+----------+-------------+ | 1 | Default | 0 | 0 | Backup | * | +--------+---------+---------+---------+----------+-------------+ *unmount The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. 3002 Device /dev/rmt/tps1d4nrvc unmounted. *label The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. Enter new Volume name: pteryx001 Defined Pools: 1: Default Item 1 selected automatically. Connecting to Storage daemon SDT-10000 at pteryx.natur.cuni.cz:9103 ... Sending label command for Volume "pteryx001" Slot 0 ... 3911 Cannot label Volume because it is already labeled: Label command failed. Do not forget to mount the drive!!! *relabel The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. Defined Pools: 1: Default Item 1 selected automatically. +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRet ention | Recycle | Slot | +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ Enter MediaId or Volume name: Volume "" has VolStatus . It must be purged before relabeling. *add You probably don't want to be using this command since it creates database records without labeling the Volumes. You probably want to use the "label" command. Defined Pools: 1: Default Item 1 selected automatically. The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. Enter number of Volumes to create. 0=>fixed name. Max=1000: 1 Enter base volume name: pteryx Enter the starting number: 1 1 Volumes created in pool Default *list *list volumes Pool: Default +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRetention | Recycle | Slot | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | 1 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 00:00:00 | 31,536,000 | 1 | 0 | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ *mount The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. 3001 Device /dev/rmt/tps1d4nrvc is mounted with Volume "" ************************************************************************* **** Since this leads to the unnamed volume I show you in the next output **** when I try to use "purge" command as suggested by "add" command: ************************************************************************* *del In general it is not a good idea to delete either a Pool or a Volume since they may contain data. You have the following choices: 1: volume 2: pool Choose catalog item to delete (1-2): 1 Defined Pools: 1: Default Item 1 selected automatically. +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRetention | Recycle | Slot | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | 1 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 00:00:00 | 31,536,000 | 1 | 0 | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ Enter MediaId or Volume name: 1 This command will delete volume pteryx0001 and all Jobs saved on that volume from the Catalog Are you sure you want to delete this Volume? (yes/no): yes *purge This command is can be DANGEROUS!!! It purges (deletes) all Files from a Job, JobId, Client or Volume; or it purges (deletes) all Jobs from a Client or Volume without regard for retention periods. Normally you should use the PRUNE command, which respects retention periods. You have the following choices: 1: files 2: jobs 3: volume Choose item to purge (1-3): 3 Defined Pools: 1: Default Item 1 selected automatically. +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRet ention | Recycle | Slot | +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ +---------+------------+-----------+-----------+----------+-------------+------- -------+---------+------+ Enter MediaId or Volume name: There are no Jobs associated with Volume . Marking it purged. Cannot purge Volume with VolStatus= sql_get.c:788 Media record for Volume "pteryx001" not found. ************************************************************************* **** So this is not the right way. Now I issued "add" again and here is **** the output of purging and labelling it again: ************************************************************************* *add You probably don't want to be using this command since it creates database records without labeling the Volumes. You probably want to use the "label" command. Defined Pools: 1: Default Item 1 selected automatically. The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. Enter number of Volumes to create. 0=>fixed name. Max=1000: 1 Enter base volume name: pteryx Enter the starting number: 1 1 Volumes created in pool Default *list volumes Pool: Default +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRetention | Recycle | Slot | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | 2 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 00:00:00 | 31,536,000 | 1 | 0 | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ *purge This command is can be DANGEROUS!!! It purges (deletes) all Files from a Job, JobId, Client or Volume; or it purges (deletes) all Jobs from a Client or Volume without regard for retention periods. Normally you should use the PRUNE command, which respects retention periods. You have the following choices: 1: files 2: jobs 3: volume Choose item to purge (1-3): 3 Defined Pools: 1: Default Item 1 selected automatically. +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRetention | Recycle | Slot | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | 2 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 00:00:00 | 31,536,000 | 1 | 0 | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ Enter MediaId or Volume name: pteryx0001 There are no Jobs associated with Volume pteryx0001. Marking it purged. *list volumes Pool: Default +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | VolRetention | Recycle | Slot | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ | 2 | pteryx0001 | DDS-4 | Purged | 0 | 1970-01-01 01:00:00 | 31,536,000 | 1 | 0 | +---------+------------+-----------+-----------+----------+--------------------- +--------------+---------+------+ *unmount The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. 3002 Device /dev/rmt/tps1d4nrvc unmounted. *label The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. Enter new Volume name: pteryx0001 Media record for new Volume "pteryx0001" already exists. *mount The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. 3001 OK mount. Device=/dev/rmt/tps1d4nrvc *run A job name must be specified. The defined Job resources are: 1: NightlySave 2: BackupCatalog 3: RestoreFiles Select Job resource (1-3): 1 Run Backup job JobName: NightlySave FileSet: Full Set Level: Incremental Client: pteryx-fd Storage: SDT-10000 Pool: Default When: 2003-07-20 14:29:53 OK to run? (yes/mod/no): yes 20-Jul-2003 14:30 pteryx-dir: Last FULL backup time not found. Doing FULL backup . 20-Jul-2003 14:30 pteryx-dir: Start Backup JobId 1, Job=NightlySave.2003-07-20_14.30.05 20-Jul-2003 14:32 pteryx-dir: NightlySave.2003-07-20_14.30.05 Error: catreq.c:34 8 Invalid Catalog request: CatReq Job=NightlySave.2003-07-20_14.30.05 GetVolInfo VolName= write=1 20-Jul-2003 14:32 pteryx-sd: NightlySave.2003-07-20_14.30.05 Warning: mount.c:230 Wanted Volume "pteryx0001". Current Volume "" not acceptable because: 1990 Invalid Catalog Request: CatReq Job=NightlySave.2003-07-20_14.30.05 GetVolInfo VolName= write=1 20-Jul-2003 14:32 pteryx-sd: Please mount Volume "pteryx0001" on Storage Device "SDT-10000" for Job NightlySave.2003-07-20_14.30.05 > It seems to me that in that version, using spaces > in the Volume name would produce errors like you are > seeing. If this is your case, just avoid using > spaces in Volume names. Use only letters and numbers > as well as the hypen or underscore. no, it is not my case. With best regards, David Komanek |
From: Kern S. <ke...@si...> - 2003-07-20 19:23:17
|
Hello, Well, it seems to me we have two variables that are different: 1. You are using an Irix compiler. 2. You are using MySQL 4.0 Concerning the compiler: if it compiles and runs, I wouldn't expect the problem to be with the compiler. Concerning MySQL: I would not be surprised to find a problem here for two reasons. 1. They have not been able to get MySQL to properly run with any gcc compiler except gcc 2.95. 2. Bacula was written for version 3.23.xx, and I imagine there will need to be changes for 4.0. At this point, I have two suggestions: 1. Either try using SQLite. It is easy to install and build, but probably not suitable for a serious=20 production environment. If it works, we know the problem is with MySQL 4.0 2. Try using MySQL version 3.23.56 (or .57) Best regards, Kern On Sun, 2003-07-20 at 19:08, David Komanek wrote: > Thank you very much for the quick response. >=20 > Yes, I use 1.30a and I am trying to get it workong on Irix 6.5.20 as both > client and server. My colleague uses it on SuSe Linux and recomended it t= o > me as a great solution for backuping. I had some problems with compiling > the source code but it seems it were only minor problems while Irix > compiler I have does not support C99 extensions. It is possible I made > some mistake in the code but the "reparations" seemed very trivial to > me .... >=20 > Well, back to the problem. I completely dropped and recreated all baculas > tables in MySQL (4.0.13). Here is the console output: >=20 > # ./bacula start > Starting the Storage daemon > Starting the File daemon > Starting the Director daemon >=20 > # console > Connecting to Director pteryx:9101 > 1000 OK: pteryx-dir Version: 1.30a (10 May 2003) > *setde > Using default Catalog name=3DMyCatalog DB=3Dbacula > Enter new debug level: 5 > Available daemons are: > 1: Director > 2: Storage > 3: Client > 4: All > Select daemon type to set debug level (1-4): 4 > Connecting to Storage daemon SDT-10000 at pteryx.natur.cuni.cz:9103 > 3000 OK setdebug=3D5 > Connecting to Client pteryx-fd at pteryx:9102 > 2000 OK setdebug=3D5 >=20 > *autodisplay on >=20 > *list volumes > Pool: Default > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | > VolRet > ention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ >=20 > *list pools > +--------+---------+---------+---------+----------+-------------+ > | PoolId | Name | NumVols | MaxVols | PoolType | LabelFormat | > +--------+---------+---------+---------+----------+-------------+ > | 1 | Default | 0 | 0 | Backup | * | > +--------+---------+---------+---------+----------+-------------+ >=20 > *unmount > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > 3002 Device /dev/rmt/tps1d4nrvc unmounted. >=20 > *label > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > Enter new Volume name: pteryx001 > Defined Pools: > 1: Default > Item 1 selected automatically. > Connecting to Storage daemon SDT-10000 at pteryx.natur.cuni.cz:9103 ... > Sending label command for Volume "pteryx001" Slot 0 ... > 3911 Cannot label Volume because it is already labeled: > Label command failed. > Do not forget to mount the drive!!! >=20 > *relabel > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > Defined Pools: > 1: Default > Item 1 selected automatically. > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | > VolRet > ention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > Enter MediaId or Volume name: > Volume "" has VolStatus . It must be purged before relabeling. >=20 > *add > You probably don't want to be using this command since it > creates database records without labeling the Volumes. > You probably want to use the "label" command. >=20 > Defined Pools: > 1: Default > Item 1 selected automatically. > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > Enter number of Volumes to create. 0=3D>fixed name. Max=3D1000: 1 > Enter base volume name: pteryx > Enter the starting number: 1 > 1 Volumes created in pool Default > *list >=20 > *list volumes > Pool: Default > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten > | VolRetention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | 1 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 > 00:00:00 > | 31,536,000 | 1 | 0 | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ >=20 > *mount > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > 3001 Device /dev/rmt/tps1d4nrvc is mounted with Volume "" >=20 >=20 >=20 > ************************************************************************* > **** Since this leads to the unnamed volume I show you in the next output > **** when I try to use "purge" command as suggested by "add" command: > ************************************************************************* >=20 > *del > In general it is not a good idea to delete either a > Pool or a Volume since they may contain data. >=20 > You have the following choices: > 1: volume > 2: pool > Choose catalog item to delete (1-2): 1 > Defined Pools: > 1: Default > Item 1 selected automatically. > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten > | VolRetention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | 1 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 > 00:00:00 > | 31,536,000 | 1 | 0 | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > Enter MediaId or Volume name: 1 >=20 > This command will delete volume pteryx0001 > and all Jobs saved on that volume from the Catalog > Are you sure you want to delete this Volume? (yes/no): yes >=20 > *purge >=20 > This command is can be DANGEROUS!!! >=20 > It purges (deletes) all Files from a Job, > JobId, Client or Volume; or it purges (deletes) > all Jobs from a Client or Volume without regard > for retention periods. Normally you should use the > PRUNE command, which respects retention periods. > You have the following choices: > 1: files > 2: jobs > 3: volume > Choose item to purge (1-3): 3 > Defined Pools: > 1: Default > Item 1 selected automatically. > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten | > VolRet > ention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > +---------+------------+-----------+-----------+----------+-------------+= ------- > -------+---------+------+ > Enter MediaId or Volume name: > There are no Jobs associated with Volume . Marking it purged. > Cannot purge Volume with VolStatus=3D > sql_get.c:788 Media record for Volume "pteryx001" not found. >=20 >=20 > ************************************************************************* > **** So this is not the right way. Now I issued "add" again and here is > **** the output of purging and labelling it again: > ************************************************************************* >=20 > *add > You probably don't want to be using this command since it > creates database records without labeling the Volumes. > You probably want to use the "label" command. >=20 > Defined Pools: > 1: Default > Item 1 selected automatically. > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > Enter number of Volumes to create. 0=3D>fixed name. Max=3D1000: 1 > Enter base volume name: pteryx > Enter the starting number: 1 > 1 Volumes created in pool Default >=20 > *list volumes > Pool: Default > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten > | VolRetention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | 2 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 > 00:00:00 > | 31,536,000 | 1 | 0 | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ >=20 > *purge >=20 > This command is can be DANGEROUS!!! >=20 > It purges (deletes) all Files from a Job, > JobId, Client or Volume; or it purges (deletes) > all Jobs from a Client or Volume without regard > for retention periods. Normally you should use the > PRUNE command, which respects retention periods. > You have the following choices: > 1: files > 2: jobs > 3: volume > Choose item to purge (1-3): 3 > Defined Pools: > 1: Default > Item 1 selected automatically. > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten > | VolRetention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | 2 | pteryx0001 | DDS-4 | Append | 0 | 0000-00-00 > 00:00:00 > | 31,536,000 | 1 | 0 | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > Enter MediaId or Volume name: pteryx0001 > There are no Jobs associated with Volume pteryx0001. Marking it purged. >=20 > *list volumes > Pool: Default > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | MediaId | VolumeName | MediaType | VolStatus | VolBytes | LastWritten > | VolRetention | Recycle | Slot | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ > | 2 | pteryx0001 | DDS-4 | Purged | 0 | 1970-01-01 > 01:00:00 > | 31,536,000 | 1 | 0 | > +---------+------------+-----------+-----------+----------+--------------= ------- > +--------------+---------+------+ >=20 > *unmount > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > 3002 Device /dev/rmt/tps1d4nrvc unmounted. >=20 > *label > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > Enter new Volume name: pteryx0001 > Media record for new Volume "pteryx0001" already exists. >=20 > *mount > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > 3001 OK mount. Device=3D/dev/rmt/tps1d4nrvc >=20 > *run > A job name must be specified. > The defined Job resources are: > 1: NightlySave > 2: BackupCatalog > 3: RestoreFiles > Select Job resource (1-3): 1 > Run Backup job > JobName: NightlySave > FileSet: Full Set > Level: Incremental > Client: pteryx-fd > Storage: SDT-10000 > Pool: Default > When: 2003-07-20 14:29:53 > OK to run? (yes/mod/no): yes >=20 >=20 > 20-Jul-2003 14:30 pteryx-dir: Last FULL backup time not found. Doing FULL= backup > . > 20-Jul-2003 14:30 pteryx-dir: Start Backup JobId 1, Job=3DNightlySave.200= 3-07-20_14.30.05 >=20 > 20-Jul-2003 14:32 pteryx-dir: NightlySave.2003-07-20_14.30.05 Error: catr= eq.c:34 > 8 Invalid Catalog request: CatReq Job=3DNightlySave.2003-07-20_14.30.05 > GetVolInfo VolName=3D write=3D1 > 20-Jul-2003 14:32 pteryx-sd: NightlySave.2003-07-20_14.30.05 Warning: > mount.c:230 Wanted Volume "pteryx0001". > Current Volume "" not acceptable because: > 1990 Invalid Catalog Request: CatReq > Job=3DNightlySave.2003-07-20_14.30.05 GetVolInfo VolName=3D write=3D1 > 20-Jul-2003 14:32 pteryx-sd: Please mount Volume "pteryx0001" on Storage = Device > "SDT-10000" for Job NightlySave.2003-07-20_14.30.05 >=20 >=20 > > It seems to me that in that version, using spaces > > in the Volume name would produce errors like you are > > seeing. If this is your case, just avoid using > > spaces in Volume names. Use only letters and numbers > > as well as the hypen or underscore. >=20 > no, it is not my case. >=20 > With best regards, >=20 > David Komanek |
From: Phil S. <al...@ca...> - 2003-07-20 19:41:16
|
On Sun, Jul 20, 2003 at 09:22:48PM +0200, Kern Sibbald wrote: > Hello, >=20 > Well, it seems to me we have two variables that are > different: >=20 > 1. You are using an Irix compiler. > 2. You are using MySQL 4.0 >=20 > Concerning the compiler: if it compiles and runs, > I wouldn't expect the problem to be with the compiler. >=20 > Concerning MySQL: I would not be surprised to find a > problem here for two reasons. 1. They have not been > able to get MySQL to properly run with any gcc compiler > except gcc 2.95. Kern, Clarification request -- I'm parsing this as "[the MySQL developers] have not been able to get MySQL [4.0] to properly run with any gcc compiler except gcc 2.95." Is this parsing correct? --=20 .********* Fight Back! It may not be just YOUR life at risk. *********. : phil stracchino : unix ronin : renaissance man : mystic zen biker geek : : al...@ca... : ala...@ea... : ph...@la... : : 2000 CBR929RR, 1991 VFR750F3 (foully murdered), 1986 VF500F (sold) : : Linux Now! ...Because friends don't let friends use Microsoft. : |
From: Kern S. <ke...@si...> - 2003-07-20 20:53:27
|
Hello Phil, On Sun, 2003-07-20 at 21:40, Phil Stracchino wrote: > On Sun, Jul 20, 2003 at 09:22:48PM +0200, Kern Sibbald wrote: > > Hello, > >=20 > > Well, it seems to me we have two variables that are > > different: > >=20 > > 1. You are using an Irix compiler. > > 2. You are using MySQL 4.0 > >=20 > > Concerning the compiler: if it compiles and runs, > > I wouldn't expect the problem to be with the compiler. > >=20 > > Concerning MySQL: I would not be surprised to find a > > problem here for two reasons. 1. They have not been > > able to get MySQL to properly run with any gcc compiler > > except gcc 2.95. >=20 > Kern, > Clarification request -- I'm parsing this as "[the MySQL developers] > have not been able to get MySQL [4.0] to properly run with any gcc > compiler except gcc 2.95." Is this parsing correct? According to their documentation, the above is true on Linux machines and for MySQL 3.23.x, and perhaps it should be nuanced by saying any gcc compiler later than 2.95. I personally have never had any problems using RedHat's 2.96 compiler to build MySQL. Best regards, Kern |
From: David K. <xd...@li...> - 2003-07-21 07:52:42
|
Thank you very much. It seems I have interesting points here: 1. it seems I resolved my original problem (and introduced another, thus :-) - my tape media was previously used by xfsdump/xfsrestore. Writing EOF at the beginning of the tape didn't help me, but completete erasing of the media did: # mt erase # ./console * label The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. Enter new Volume name: pteryx0001 Defined Pools: 1: Default Item 1 selected automatically. Connecting to Storage daemon SDT-10000 at pteryx.natur.cuni.cz:9103 ... Sending label command for Volume "pteryx0001" Slot 0 ... mount 3000 OK label. Volume=pteryx0001 Device=/dev/rmt/tps1d4nrvc Catalog record for Volume "pteryx0001", Slot 0 successfully created. Requesting mount SDT-10000 ... 3001 Mounted Volume: pteryx0001 3001 Device /dev/rmt/tps1d4nrvc is mounted with Volume "pteryx0001" **The defined Storage resources are: 1: SDT-10000 Item 1 selected automatically. 3001 Device /dev/rmt/tps1d4nrvc is mounted with Volume "pteryx0001" * 21-Jul-2003 01:05 pteryx-dir: Last FULL backup time not found. Doing FULL backup . 21-Jul-2003 01:05 pteryx-dir: Start Backup JobId 3, Job=NightlySave.2003-07-21_0 1.05.00 21-Jul-2003 01:05 pteryx-sd: Wrote label to prelabeled Volume "pteryx0001" on de vice /dev/rmt/tps1d4nrvc * 21-Jul-2003 04:10 pteryx-dir: Bacula 1.30a (10May03): 21-Jul-2003 04:10 JobId: 3 Job: NightlySave.2003-07-21_01.05.00 FileSet: Full Set Backup Level: Full Client: pteryx-fd Start time: 21-Jul-2003 01:05 End time: 21-Jul-2003 04:10 Files Written: 205,137 Bytes Written: 14,958,906,020 Rate: 1341.8 KB/s Software Compression: None Volume names(s): pteryx0001 Volume Session Id: 1 Volume Session Time: 1058721009 Last Volume Bytes: 15,007,075,383 FD termination status: OK SD termination status: OK Termination: Backup OK 21-Jul-2003 04:10 pteryx-dir: Begin pruning Jobs. 21-Jul-2003 04:10 pteryx-dir: No Jobs found to prune. 21-Jul-2003 04:10 pteryx-dir: Begin pruning Files. 21-Jul-2003 04:10 pteryx-dir: No Files found to prune. 21-Jul-2003 04:10 pteryx-dir: End auto prune. 2. It seems my further problem is to RunBefore scripts is not passed the password of Bacula's MzSQL user. But this is not related to my previous problem and should be solvable quite easy: 21-Jul-2003 04:10 pteryx-dir: RunBefore: /usr/local/bin/mysqldump: Got error: 10 45: Access denied for user: 'bacula@localhost' (Using password: NO) when trying to connect 21-Jul-2003 04:10 pteryx-dir: Start Backup JobId 4, Job=BackupCatalog.2003-07-21 _01.10.00 21-Jul-2003 04:10 pteryx-dir: Bacula 1.30a (10May03): 21-Jul-2003 04:10 JobId: 4 Job: BackupCatalog.2003-07-21_01.10.00 FileSet: Catalog Backup Level: Full Client: pteryx-fd Start time: 21-Jul-2003 04:10 End time: 21-Jul-2003 04:10 Files Written: 1 Bytes Written: 95 Rate: 0.0 KB/s Software Compression: None Volume names(s): pteryx0001 Volume Session Id: 2 Volume Session Time: 1058721009 Last Volume Bytes: 15,007,075,888 FD termination status: OK SD termination status: OK Termination: Backup OK 21-Jul-2003 04:10 pteryx-dir: Begin pruning Jobs. 21-Jul-2003 04:10 pteryx-dir: No Jobs found to prune. 21-Jul-2003 04:10 pteryx-dir: Begin pruning Files. 21-Jul-2003 04:10 pteryx-dir: No Files found to prune. 21-Jul-2003 04:10 pteryx-dir: End auto prune. 3. As regards to MySQL/GCC discussion: I use GCC 3.2.2 from freeware.sgi.com, i.e. GNU/GPL compiler prepared by the vendor of my machine :-) I compiled MySQL 4.0.13 with the folowing options (using innodb): CC="gcc -mabi=64" CXX="c++ -mabi=64" CFLAGS="-I/usr/local/include -I/usr/include " LDFLAGS="-L/usr/local/lib -L/usr/lib64" ./configure --prefix=/usr/local \ --enable-thread-safe-client \ --with-pthread \ --with-libwrap \ --without-debug \ --with-openssl \ --without-bench \ --with-charset=win1250 \ --with-extra-charsets=latin1,latin2,latin5,cp1251,win1251ukr,koi8_ru There were only two problems: (1) conflicting curses.h and term.h in mysql.cc, line 54 and 55, (2) tcpwrappers support doesn't seem to work properly in mysqld. Best regards, David Komanek |
From: Kern S. <ke...@si...> - 2003-07-21 09:09:07
|
Hello, On Mon, 2003-07-21 at 09:52, David Komanek wrote: > Thank you very much. >=20 > It seems I have interesting points here: >=20 > 1. it seems I resolved my original problem (and introduced another, thus > :-) - my tape media was previously used by xfsdump/xfsrestore. Writing > EOF at the beginning of the tape didn't help me, but completete erasing o= f > the media did: Very interesting. I'll remember this for the next time. >=20 > # mt erase > # ./console > * label > The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > Enter new Volume name: pteryx0001 > Defined Pools: > 1: Default > Item 1 selected automatically. > Connecting to Storage daemon SDT-10000 at pteryx.natur.cuni.cz:9103 ... > Sending label command for Volume "pteryx0001" Slot 0 ... >=20 > mount > 3000 OK label. Volume=3Dpteryx0001 Device=3D/dev/rmt/tps1d4nrvc > Catalog record for Volume "pteryx0001", Slot 0 successfully created. > Requesting mount SDT-10000 ... > 3001 Mounted Volume: pteryx0001 > 3001 Device /dev/rmt/tps1d4nrvc is mounted with Volume "pteryx0001" > **The defined Storage resources are: > 1: SDT-10000 > Item 1 selected automatically. > 3001 Device /dev/rmt/tps1d4nrvc is mounted with Volume "pteryx0001" > * > 21-Jul-2003 01:05 pteryx-dir: Last FULL backup time not found. Doing FULL > backup > . > 21-Jul-2003 01:05 pteryx-dir: Start Backup JobId 3, > Job=3DNightlySave.2003-07-21_0 > 1.05.00 > 21-Jul-2003 01:05 pteryx-sd: Wrote label to prelabeled Volume "pteryx0001= " > on de > vice /dev/rmt/tps1d4nrvc > * > 21-Jul-2003 04:10 pteryx-dir: Bacula 1.30a (10May03): 21-Jul-2003 04:10 > JobId: 3 > Job: NightlySave.2003-07-21_01.05.00 > FileSet: Full Set > Backup Level: Full > Client: pteryx-fd > Start time: 21-Jul-2003 01:05 > End time: 21-Jul-2003 04:10 > Files Written: 205,137 > Bytes Written: 14,958,906,020 > Rate: 1341.8 KB/s > Software Compression: None > Volume names(s): pteryx0001 > Volume Session Id: 1 > Volume Session Time: 1058721009 > Last Volume Bytes: 15,007,075,383 > FD termination status: OK > SD termination status: OK > Termination: Backup OK >=20 > 21-Jul-2003 04:10 pteryx-dir: Begin pruning Jobs. > 21-Jul-2003 04:10 pteryx-dir: No Jobs found to prune. > 21-Jul-2003 04:10 pteryx-dir: Begin pruning Files. > 21-Jul-2003 04:10 pteryx-dir: No Files found to prune. > 21-Jul-2003 04:10 pteryx-dir: End auto prune. >=20 >=20 >=20 >=20 > 2. It seems my further problem is to RunBefore scripts is not passed the > password of Bacula's MzSQL user. But this is not related to my previous > problem and should be solvable quite easy: No, Bacula does not pass any password. You'll need to put it in your script or create a script since it appears you were calling mysqldump directly. >=20 > 21-Jul-2003 04:10 pteryx-dir: RunBefore: /usr/local/bin/mysqldump: Got > error: 10 > 45: Access denied for user: 'bacula@localhost' (Using password: NO) when > trying > to connect > 21-Jul-2003 04:10 pteryx-dir: Start Backup JobId 4, > Job=3DBackupCatalog.2003-07-21 > _01.10.00 > 21-Jul-2003 04:10 pteryx-dir: Bacula 1.30a (10May03): 21-Jul-2003 04:10 > JobId: 4 > Job: BackupCatalog.2003-07-21_01.10.00 > FileSet: Catalog > Backup Level: Full > Client: pteryx-fd > Start time: 21-Jul-2003 04:10 > End time: 21-Jul-2003 04:10 > Files Written: 1 > Bytes Written: 95 > Rate: 0.0 KB/s > Software Compression: None > Volume names(s): pteryx0001 > Volume Session Id: 2 > Volume Session Time: 1058721009 > Last Volume Bytes: 15,007,075,888 > FD termination status: OK > SD termination status: OK > Termination: Backup OK >=20 > 21-Jul-2003 04:10 pteryx-dir: Begin pruning Jobs. > 21-Jul-2003 04:10 pteryx-dir: No Jobs found to prune. > 21-Jul-2003 04:10 pteryx-dir: Begin pruning Files. > 21-Jul-2003 04:10 pteryx-dir: No Files found to prune. > 21-Jul-2003 04:10 pteryx-dir: End auto prune. >=20 >=20 >=20 > 3. As regards to MySQL/GCC discussion: I use GCC 3.2.2 from > freeware.sgi.com, i.e. GNU/GPL compiler prepared by the vendor of my > machine :-) I compiled MySQL 4.0.13 with the folowing options (using > innodb): OK, good luck. >=20 > CC=3D"gcc -mabi=3D64" CXX=3D"c++ -mabi=3D64" CFLAGS=3D"-I/usr/local/inclu= de -I/usr/include > " LDFLAGS=3D"-L/usr/local/lib -L/usr/lib64" ./configure --prefix=3D/usr/l= ocal \ > --enable-thread-safe-client \ > --with-pthread \ > --with-libwrap \ > --without-debug \ > --with-openssl \ > --without-bench \ > --with-charset=3Dwin1250 \ > --with-extra-charsets=3Dlatin1,latin2,latin5,cp1251,win1251ukr,koi8_ru >=20 > There were only two problems: (1) conflicting curses.h and term.h in > mysql.cc, line 54 and 55, (2) tcpwrappers support doesn't seem to work > properly in mysqld. >=20 > Best regards, >=20 > David Komanek |
From: Jari F. <ja...@bi...> - 2003-07-20 19:45:54
|
>Well, it seems to me we have two variables that are >different: > >1. You are using an Irix compiler. >2. You are using MySQL 4.0 > >Concerning the compiler: if it compiles and runs, >I wouldn't expect the problem to be with the compiler. > >Concerning MySQL: I would not be surprised to find a >problem here for two reasons. 1. They have not been Hmm.. I'm hijacking the thread now.. But just came to my mind. I converted by bacula database to MySQL 4.x, with and especially using Innobase. MySQL itself did't complaint, but InnoDB - like MOST databases DO NOT like using BLOB's in indices! No bloblem, I converted all filename etc. types to strings (varchar) of some reasonable length, and no problem. No my MySQL 4.x is running with InnoDB tables, and the performance is propably - if not notieably ;) better. My suggestion is here, that bacula didn't use BLOB's - at all. They are a very special data type, and I could understand if the backed up file would be stored to the DB.. then it would be a BLOB ;) It's portability and performance issue. Regards, jarif |
From: Kern S. <ke...@si...> - 2003-07-21 13:11:41
|
Hello, On Sun, 2003-07-20 at 21:45, Jari Fredriksson wrote: > >Well, it seems to me we have two variables that are > >different: > > > >1. You are using an Irix compiler. > >2. You are using MySQL 4.0 > > > >Concerning the compiler: if it compiles and runs, > >I wouldn't expect the problem to be with the compiler. > > > >Concerning MySQL: I would not be surprised to find a > >problem here for two reasons. 1. They have not been >=20 >=20 > Hmm.. I'm hijacking the thread now.. But just came to my mind. >=20 > I converted by bacula database to MySQL 4.x, with and especially using > Innobase. >=20 > MySQL itself did't complaint, but InnoDB - like MOST databases DO NOT lik= e > using BLOB's in indices! >=20 > No bloblem, I converted all filename etc. types to strings (varchar) of s= ome > reasonable length, and no problem. No my MySQL 4.x is running with InnoDB > tables, and the performance is propably - if not notieably ;) better. >=20 > My suggestion is here, that bacula didn't use BLOB's - at all. They are a > very special data type, and I could understand if the backed up file woul= d > be stored to the DB.. then it would be a BLOB ;) This is undoubtedly true, but I know next to nothing about SQL databases, and making any changes would generally disrupt a lot of users, so I am obliged to leave it unchanged. Perhaps a note in the manual on this would be good though. >=20 > It's portability and performance issue. Best regards, Kern |
From: Phil S. <al...@ca...> - 2003-07-20 21:43:12
|
On Sun, Jul 20, 2003 at 10:52:53PM +0200, Kern Sibbald wrote: > On Sun, 2003-07-20 at 21:40, Phil Stracchino wrote: > > On Sun, Jul 20, 2003 at 09:22:48PM +0200, Kern Sibbald wrote: > > > Concerning MySQL: I would not be surprised to find a > > > problem here for two reasons. 1. They have not been > > > able to get MySQL to properly run with any gcc compiler > > > except gcc 2.95. > > Kern, > > Clarification request -- I'm parsing this as "[the MySQL developers] > > have not been able to get MySQL [4.0] to properly run with any gcc > > compiler except gcc 2.95." Is this parsing correct? > According to their documentation, the above is true on Linux > machines and for MySQL 3.23.x, and perhaps it should be nuanced > by saying any gcc compiler later than 2.95. >=20 > I personally have never had any problems using RedHat's > 2.96 compiler to build MySQL. Ah. I was wondering, because I've built several MySQL-3.23 versions on Linux using GCC-3.0 up to 3.3. If the're saying gcc-2.95 *or later*, that's a rather different thing. :) --=20 .********* Fight Back! It may not be just YOUR life at risk. *********. : phil stracchino : unix ronin : renaissance man : mystic zen biker geek : : al...@ca... : ala...@ea... : ph...@la... : : 2000 CBR929RR, 1991 VFR750F3 (foully murdered), 1986 VF500F (sold) : : Linux Now! ...Because friends don't let friends use Microsoft. : |
From: Kern S. <ke...@si...> - 2003-07-21 13:22:12
|
Hello Phil, I see they have updated their warning since I last looked at it: Compiler Advisory: Several users have reported random crashes and table corruptions when using MySQL binaries compiled with gcc 2.96 on the x86 Linux platform. We suggest that you use gcc 2.95 or gcc 2.91 to compile your own binaries. It should also be safe to use gcc 3.2. Found at: http://www.mysql.com/downloads/mysql-3.23.html So, it looks like gcc 3.2 is good as well. The same warning also appears in the 4.0 download section. Best regards, Kern On Sun, 2003-07-20 at 23:39, Phil Stracchino wrote: > On Sun, Jul 20, 2003 at 10:52:53PM +0200, Kern Sibbald wrote: > > On Sun, 2003-07-20 at 21:40, Phil Stracchino wrote: > > > On Sun, Jul 20, 2003 at 09:22:48PM +0200, Kern Sibbald wrote: > > > > Concerning MySQL: I would not be surprised to find a > > > > problem here for two reasons. 1. They have not been > > > > able to get MySQL to properly run with any gcc compiler > > > > except gcc 2.95. >=20 > > > Kern, > > > Clarification request -- I'm parsing this as "[the MySQL developers] > > > have not been able to get MySQL [4.0] to properly run with any gcc > > > compiler except gcc 2.95." Is this parsing correct? >=20 > > According to their documentation, the above is true on Linux > > machines and for MySQL 3.23.x, and perhaps it should be nuanced > > by saying any gcc compiler later than 2.95. > >=20 > > I personally have never had any problems using RedHat's > > 2.96 compiler to build MySQL. >=20 > Ah. I was wondering, because I've built several MySQL-3.23 versions on > Linux using GCC-3.0 up to 3.3. If the're saying gcc-2.95 *or later*, > that's a rather different thing. :) >=20 >=20 |
From: David K. <xd...@li...> - 2003-07-21 08:21:57
|
Hello all, as I currently try to solve my problems using Bacula, two more questions emerged: 1. Is there anybody interested in porting Bacula server daemons on Irix ? It seems my instances of daemons are quite working now, so I could try to solve the compiler issues, if there would be interest for it. 2. [probably for bacula-users list, but I hope somebody could tell me here too] Is there anz stronger way of data integrity test on tapes than use of "bls devicename" ? If I understand correctly the manual, this tool only lists contents of the tape but does not ensure that every bit of the written data is readable. In addition to this, what is the best strategy to automate some kind of verification of the data integrity on tapes written by bacula ? Sincerely, David Komanek |
From: Kern S. <ke...@si...> - 2003-07-21 09:25:58
|
Hello, On Mon, 2003-07-21 at 10:21, David Komanek wrote: > Hello all, >=20 > as I currently try to solve my problems using Bacula, two more questions > emerged: >=20 > 1. Is there anybody interested in porting Bacula server daemons on Irix ? > It seems my instances of daemons are quite working now, so I could try to > solve the compiler issues, if there would be interest for it. Bacula should compile and run on Irix. Version 1.30a may have a few minor errors that need fixing. The current BETA version of 1.31 should compile without errors at least on Irix 6.5. >=20 > 2. [probably for bacula-users list, but I hope somebody could tell me > here too] Is there anz stronger way of data integrity test on tapes than > use of "bls devicename" ? If I understand correctly the manual, this tool > only lists contents of the tape but does not ensure that every bit of the > written data is readable. In addition to this, what is the best strategy > to automate some kind of verification of the data integrity on tapes > written by bacula ? The answer depends on what you are trying to accomplish. If you just want to test it one time, then the best bet is to do a restore and use diff to ensure everything was put back correctly (in a different directory tree). If you want to always verify your tapes then you should run a verify job with level VerifyToCatalog. This will compare the attributes on the tape to what is in the catalog, and in doing so, it will read through=20 all the data for the job. Some people have reported problems with this option in 1.30a, so I have modified the code in 1.31, where I am sure it works. See: http://www.bacula.org/html-manual/dirdconf.html#JobResource Best regards, Kern |