El martes, 3 de julio de 2018, 8:55:32 (UTC-3), Alejandro Hamann escribió: > Hello BareOS community. > > I have backup jobs that end with status OK with warning. > > The result of messages command of bconsole showme that that I have a couple > of SD Errors. But I want to know more specific information about those SD > Errors. I've see in /var/log/bareos/ and there's nothing there (apart from > bareos-audit.log.* and bareos.log.* ). > > When running "dmesg -T" and cross dates and times of BareOS problematic jobs, > I can get some information that might be useful. > But in reality, I would like to know if there is anything in BareOS that > gives me more information than just a problem counter. > > Regards! > > -- > Pablo Alejandro Hamann
Hello Jörg Steffens. Thank you for your response (and sorry for the delay in thanking, I'm just getting a little updated with our implementation of backup solution, after a long pause). I read all the documentation of Bareos (very long and complete by the way), which always in one way or another, covers almost all the necessary aspects to have the system in operation. Now, according to the documentation, when I get a system message like: st 0: 0: 0: 0: [st0] Failed to read 64512 byte block with 80 byte transfer. that means that, the block was written with 64512 block size but we only read 80 bytes (exactly the opposite of what you described to me). So, the label would be fine (64K which is the desired size and default), and the problem would resides when the tape unit tries to read the label. However, when I did the tests to determine the maximum possible speed (using btape, as described in the manual and reproducing the tests that are discussed in Bareos Whitepaper Tape Speed Tuning), I had no problem. dmesg simply did not report any errors. By the way, in my configuration I use 512k blocksizes because it is the biggest thing that my system supports (if I use bigger blocks, like 1 MB, something fails, I do not know if the tapedrive, the SAS controller, or any other hardware component) . the system is a GNU / Linux Debian Stretch 9.5 (using the Proxmox kernels, although if I use the Debian kernels, that does not change at all). The hardware is one of those small Dell PowerEdge T30 Servers (equipped with an Intel Xeon E3-1225 v5 @ 3.30GHz and 8GB of ECC RAM) and an HP Smart Array P411 SAS controller, and (obviously) the tape library that is an IBM TS3100 LTO-5. Even more, every time I make a tape backup, I get these errors, which causes the backup jobs to end with warnings (instead of OK); however, nothing fails, I can restore perfectly and without problems (although dmesg keeps reporting that Failed to read 64512 byte block with 80 byte transfer). What else can I do with these error messages? (It seems that in the end they are just that: error messages, but not errors in themselves). I have never touched the mt configuration. It is the default one in Debian. I understand that it uses variable block size by default (otherwise, all tests with btape would have failed me). A little light please :-/ Regards! -- You received this message because you are subscribed to the Google Groups "bareos-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. For more options, visit https://groups.google.com/d/optout.
