 |
Is this a local (on the same sd) or a remote (to a different sd) copy? Can you please try of the issue still exists in our nightly build https://download.bareos.org/bareos/experimental/nightly/ |
|
|
It's local, on same sd. Will try nightly and report back. Thanks. |
|
|
No difference with nightly Version. 09-Jul 16:31 bareos-dir JobId 431: Copying using JobId=413 Job=asterix-lka-ndmp-job.2019-07-07_09.00.00_16 09-Jul 16:31 bareos-dir JobId 431: Bootstrap records written to /var/lib/bareos/bareos-dir.restore.1.bsr 09-Jul 16:31 bareos-dir JobId 431: Start Copying JobId 431, Job=copy-asterix-lka-copypool.2019-07-09_16.31.30_10 09-Jul 16:31 bareos-dir JobId 431: Fatal error: Connect failure: ERR=error:1408F10B:SSL routines:ssl3_get_record:wrong version number 09-Jul 16:31 bareos-dir JobId 431: Error: TLS shutdown failure.: ERR=error:140E0197:SSL routines:SSL_shutdown:shutdown while in init 09-Jul 16:31 bareos-dir JobId 431: Fatal error: TLS negotiation failed 09-Jul 16:31 bareos-dir JobId 431: Error: Bareos bareos-dir 19.1.2 (01Feb19): Build OS: Linux-4.4.175-89-default debian Debian GNU/Linux 9.9 (stretch) Prev Backup JobId: 413 Prev Backup Job: asterix-lka-ndmp-job.2019-07-07_09.00.00_16 New Backup JobId: 432 Current JobId: 431 Current Job: copy-asterix-lka-copypool.2019-07-09_16.31.30_10 Backup Level: Full Client: asterix-ndmp FileSet: "asterix-lka-ndmp-fs" Read Pool: "NDMPTape" (From Job resource) Read Storage: "NDMPTape" (From Pool resource) Write Pool: "NDMPCopy" (From Job Pool's NextPool resource) Write Storage: "NDMPCopy" (From Storage from Pool's NextPool resource) Next Pool: "NDMPCopy" (From Job Pool's NextPool resource) Catalog: "MyCatalog" (From Default catalog) Start time: 09-Jul-2019 16:31:32 End time: 09-Jul-2019 16:31:32 Elapsed time: 0 secs Priority: 10 SD Files Written: 0 SD Bytes Written: 0 (0 B) Rate: 0.0 KB/s Volume name(s): Volume Session Id: 0 Volume Session Time: 0 Last Volume Bytes: 0 (0 B) SD Errors: 0 SD termination status: Bareos binary info: bareos.org build: Get official binaries and vendor support on bareos.com Termination: *** Copying Error *** |
|
|
That's weird. Our testsuite should have tested this before release. Can you provide your storage configuration for the two storages and paired storages, so I can test this? |
|
|
Storage { Name = NDMPCopy Address = 10.210.0.88 # N.B. Use a fully qualified name here Port = 10000 Protocol = NDMPv4 # Need to specify protocol before password as protocol determines password encoding used. Auth Type = Clear # Clear == Clear Text, MD5 == Challenge protocol Username = ndmp # username of the NDMP user on the TAPE AGENT e.g. the Bareos SD but accessed via the NDMP protocol. Password = xxxxxed # password of the NDMP user on the TAPE AGENT e.g. the Bareos SD but accessed via the NDMP protocol. Device = scalar-i500 Media Type = LTO-5 PairedStorage = i500 Maximum Concurrent Jobs = 6 } Storage { Name = NDMPTape Address = 10.210.0.88 # N.B. Use a fully qualified name here Port = 10000 Protocol = NDMPv4 # Need to specify protocol before password as protocol determines password encoding used. Auth Type = Clear # Clear == Clear Text, MD5 == Challenge protocol Username = ndmp # username of the NDMP user on the TAPE AGENT e.g. the Bareos SD but accessed via the NDMP protocol. Password = xxxxxed # password of the NDMP user on the TAPE AGENT e.g. the Bareos SD but accessed via the NDMP protocol. Device = scalar-i500 Media Type = LTO-5 PairedStorage = i500 Maximum Concurrent Jobs = 6 } Storage { Name = i500 Address = bb8.elkb.de # N.B. Use a fully qualified name here (do not use "localhost" here). Password = xxxxxxed Device = scalar-i500 Media Type = LTO-5 AutoChanger = yes Maximum Concurrent Jobs = 6 } |
|
|
OK, If I look at it now, I will change the IP Adresses into names and test again. |
|
|
As far as I can tell you should have only two storages: the Bareos storage "i500" and *one* paired NDMP storage. I think the second NDMP storage leads to confusion inside of Bareos. |
|
|
When I try to copy my Tape Backups to a copy Tape to tranfer to vault I get this: 12-Jul 06:22 bareos-dir JobId 452: Fatal error: JobId 413 cannot Copying using the same read and write storage. 12-Jul 06:22 bareos-dir JobId 452: Error: Bareos bareos-dir 19.1.2 (01Feb19): Thats why I added a second NDMP Storage in the first place. I tested with a i500-copy paired to NDMPCopy. It gives also the TLS Error. So it looks like its not possible at the moment to have a NDMP Copy Pool ? Or am I doing it wrong ? |
|
|
Our Testsuite uses the following: Configure two NDMP Storages in director (copy source and copy target) each with its own Bareos Storage as Paired Storage. The Bareos Storages may point to the same Device on the same SD. This is what our automated test does. As 18.2.5 has of course been tested with the testsuite before release, I'm quite sure this is some kind of configuration issue (and maybe a lack of proper documentation for NDMP). |
|
|
Thanks for your time and your support. I will try as you suggested. Case can be closed at your convenience. |
|
|
I'd really appreciate your feedback whether you got it working or not, so we know what to add to the documentation exactly. |
|
|
Can't get this to work. If I disable TLS I get different error Messages like: 15-Jul 08:27 bareos-dir JobId 531: Fatal error: Password encoding is not MD5. You are probably restoring a NDMP Backup with a restore job not configured for NDMP protocol. 15-Jul 08:27 bareos-dir JobId 531: Fatal error: Director unable to authenticate with Storage daemon at "bb8-storage.elkb.de:10000". Possible causes: Passwords or names not the same or TLS negotiation problem or Maximum Concurrent Jobs exceeded on the SD or SD networking messed up (restart daemon). Changed encoding to MD5, double checked Username and Password and Maximum Concurrent Jobs. Have to give up for now. |
|
|
Could you also provide your Pool, Job (and if applicable JobDefs) configuration? This should be simple and work out of the box. |
|
|
sure: Pool { Name = NDMPTape Pool Type = Backup Recycle = yes AutoPrune = yes Storage = NDMPTape Recycle Pool = Scratch Maximum Block Size = 1048576 Volume Retention = 6 Month Next Pool = NDMPCopy } Pool { Name = NDMPCopy Pool Type = Backup Recycle = yes AutoPrune = yes Storage = NDMPCopy Recycle Pool = Scratch Maximum Block Size = 1048576 Volume Retention = 6 Month } Job { Name = copy-asterix-lka-copypool client = None Type = Copy Messages = Standard Pool = NDMPTape File Set = None Schedule = sun-22 Selection Type = SQL Query Selection Pattern = " SELECT MAX(jobid) FROM job WHERE name='asterix-lka-ndmp-job' AND type='B' AND level='F' AND jobstatus='T'; " Enabled = no } # Fake client for copy jobs Client { Name = None Address = localhost Protocol = NDMPv4 Auth Type = Clear Username = ndmp-copy Password = xxxxed Catalog = MyCatalog Enabled = yes } # Fake fileset for copy jobs Fileset { Name = None Include { Options { } } } # ndmp copy agent in bareos-sd.d/ndmp Ndmp { Name = None Username = ndmp-copy Password = xxxxed AuthType = Clear } |
|
|
A copy Job doesn't require a Client, so you can remove the directive. If you want to keep the dummy, remove the NDMP settings. Even though this is an NDMP Backup copying it is not a NDMP operation. |
|
|
>> Even though this is an NDMP Backup copying it is not a NDMP operation. << This was the right hint !!! If it's not NDMP, the Copy Storage must look like this: Storage { Name = NDMPCopy Address = bb8.elkb.de # N.B. Use a fully qualified name here Password = "xxxxxed" Device = scalar-i500 Media Type = LTO-5 PairedStorage = i500-copy Maximum Concurrent Jobs = 6 } You made my day !!! It's loading Tapes and copying Blocks. So Cool, thank you very much. |
|
|
One don'T even need a different Paired Storage, it also works with: PairedStorage = i500 |
|
|
Do you have any idea what we could add to the documentation so nobody else runs into this issue? |
|
|
I was so focused on NDMP, that for me it was not clear that copying an NDMP Backup is not an NDMP Operation and that the Copy Storage is just a "normal" Bareos Storage. No Need for a Fake Client and a Fake Fileset, and no need for a 2nd ndmp_tape_agent. No NDMP at all involved here. |
|
|
Added note to NDMP documentation so this case becomes clearer. |
|
|
Fix committed to bareos master branch with changesetid 11595. |
- Anonymous
