Jump to content

merrisr

Members
  • Content count

    11
  • Joined

  • Last visited

Community Reputation

0 Neutral

About merrisr

  • Rank
    Occasional Forum Poster
  1. Okay, I applied the 4.1 driver update and the .319 Retrospect server updates, still having the same issue.
  2. Hello, I have Retrospect Multiserver 6.5.276 with all options (Multi Server Value Package) running on a Windows 2000 server with 3 Ecrix (now bought out by Exabyte) Autopak 2x15 (two drives, 15-slot library) autoloaders connected to an Adaptec 39160 Ultra 160 SCSI contoller. I have applied the firmware updates to the drives, but there is no flashable library firmware. In default configuration the libraries actually show up as "Spectra Logic Library" and the drives show up as "Ecrix VXA DC". Anyway, everything works fine, except that the libraries sometimes disappear from the Storage Devices Status screen. Then, automated backup operations can't find the tapes they need. Going to the Environment tab and hitting refresh usually clears this up, and then the backup operation can continue, but that requires frequent monitoring and intervention. I'll be happy to email a jpeg of what I'm talking about if anyone is interested. merrisr@saic.com
  3. Ah, no, I left it alone for over a week and it didn't time out. But, killing it with pskill and restarting it with psexec has done the trick on all three systems I tried, so I'm good. Thanks, Rhian
  4. Thanks for the quick responses and the link. Those command-line options should be helpful. So, great news - I tried Sysinternals' freeware pstools (pskill and psexec) and killed the retroclient process remotely, then restarted it and that worked! I'm not sure why just restarting the service didn't work, but maybe that had to do with leaving it idle. Did you mean to leave it idle after stopping and before restarting? If so, I didn't really do that, so maybe that made a difference. Thanks! I'll post more if I discover anything else relevant, but I'm in much better shape if I can at least kill and restart the client. Rhian Merris merrisr@saic.com
  5. But that doesn't clear it up in the Windows boxes - can't seem to clear it without rebooting the client completely.
  6. merrisr

    Memory Issues?

    By the way, to add to my last post, AmyC, I'm not saying you're wrong. In fact the errors in the execution log (TMemory::mhalloc: VirtualAlloc (104.5 M, MEM_RESERVE) failed, error 8) look like they bear out what you said. I just don't understand why Retrospect isn't using more of the swap file. I'm not saying that's necessarily a retrospect bug as opposed to a windows bug, but I don't care for it either way. Rhian Merris merrisr@saic.com
  7. I am running Retrospect 6.5 multiserver, backing up approximately 30 clients, c. 25 Windows and c. 5 linux. I have had a number of Retrospect server crashes, after which some of the clients cannot be accessed. The error is error -505 (backup client reserved). The User's Guide states to resolve this by rebooting both systems, but some of the client systems involved are 24/7 or close to it, so I cannot always reboot them when I need to do so. This is a big problem. There are a ton of posts and resources for dealing with this issue in OS/X, and I can easily resolve it in linux by killing the client processes and manually restarting, but I cannot resolve it for my windows clients without rebooting. I have restarted the services, etc., but no luck without rebooting. Does anyone know of a workaround short of rebooting the entire system? If not, this poses a big problem to utilization of Retrospect in a high-availability environment. Thanks, Rhian Merris merrisr@saic.com
  8. I've run into several similar or related issues. I don't think that the CPU utilization is the problem - my linux clients peg at 99+% utilization even when no backup is running (though I've resolved that by setting their nice to 19). But the client reserved issue is very annoying. I've gotten that with linux and windows clients, and the user's guide and Dantz tech support both say the same thing - you have to reboot both systems. Well, that's not cool since I'm backing up a number of 24/7-type boxes, so I can't just go around rebooting them whenever I want. With linux that's not a big deal - I can just kill the processes and manually restart the client, but I haven't figured out how to clear the reserved issue for windows machines - restarting the service doesn't do it. Rhian Merris merrisr@saic.com
  9. merrisr

    Linux client CPU usage

    I've seen the same thing with 6.0 and 6.5. One thing you can try is setting the nice level for the retrospect client to something high - I set one of mine to 19 and that dropped utilization from 99% to 0. The client system had 2 CPUs, and one of them was pegged at 99% even when it wasn't being backed up. Of course, you'll want to test and make sure your backups are running properly after niceing them. Rhian Merris merrisr@saic.com
  10. merrisr

    Memory Issues?

    I don't understand that - I have a server with 1GB real memory and 4GB swap file that gets those same errors backing up large linux systems, with peak usage at about 830MB - i.e., using up all the real memory that the OS, etc. are not using, but not taking advantage of the swap file. Is there any way to get around that? Rhian Merris merrisr@saic.com
×