Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by shadowspawn

  1. On my VXA setup connected via an Adaptec 2930 (see earlier posting in this thread for details), I was beginning to despair after getting almost nothing backed up five nights out of six. Retrospect Server hung or stalled twice and did nothing productive overnight, two hardware errors early in night, switched server to another computer but hardware error from tape unit in first five minutes. I then reconnected the same Ecrix VXA tape drive to the computer using a MicroTech SCIS/firewire adaptor instead of directly via SCSI. I then ran Retrospect for over 48 hours almost continuously backing up network clients to tape. One apparent hang while I was playing about in the log during a backup, one potential stall when a Net Retry dialog came up. In both cases since I was present, backup was underway again in under five minutes. It is too early to say for sure that things are really better bypassing the SCSI card, I may just have hit a good patch statistically speaking. As noted earlier however, my problems have not been due to major stalls on 200 MB boundaries, but due to the backup stalling before backing up anything off some clients and some hardware errors. I have not looked at performance yet to see if it has dropped, I am trying to get a backup at all before I worry about the speed!
  2. On my second VXA setup I am getting hangs and stalls during client backups, but perhaps not the same problem others are seeing here. My problem most often occurs before Retrospect shows any files being copied from the client. This might be when Retrospect starts positioning the tape or while communicating with the client, I am not sure yet. Sometimes Retrospect is still responding although has achieved nothing in over 12 hours, sometimes it is burning CPU and must be killed. Each time so far it has been a Mac OS 9 client, although I have had a few clueless crashes, and a couple of hardware errors to muddy the waters too. I have managed to backup about 130 GB so far, but about every second night something goes wrong, and I am still trying to identify the points of failure. Running Retrospect Server 5.0.203 on 466 MHz G4 running Mac OS X 10.1.3. Adaptec 2930U 50-pin SCSI card, 1.1 driver. VAX-1e SCSI drive, Retrospect reports Ecrix VXA-1 Version 2959, Ecrix VXA DC (4.08) Backing up 40 clients, including Mac OS 9, (5.0.201 clients) X, X Server, (5.0.528 clients) Windows 98 SE, 2000, XP (5.0.112 clients)
  3. Prompted by Dave laying down the smack about supplying system details... :-) One of my VXA setups has now done 27 GB without problems. Two full backups, two incrementals. I have not noticed pauses during the backups, but I usually run the backups unattended. VXA Model VXA-1e Narrow SCSI from VXA unit to MicroTech firewire adaptor (as supplied by Ecrix). Retrospect Desktop 5.0.203. Retrospect Desktop computer is dual 450 MHz G4 running Mac OS X, 10.1.3. One network client on Mac OS 9. Retrospect says this about the device: ID:FireWire-A, Vendor:Ecrix, Product:VXA-1, Version:2848, Driver:Ecrix VXA DC (4.08) and Apple System Profiler says this: VXA-1 2848 Vendor ID: d016 Speed: 200 Mb/sec GUID: 13637120d Vendor name: ECRIX Unit version: 66691 I don't know how to get SCSI to firewire adaptor firmware information?
  4. I narrowed down my crash to backing up one particular volume. Following a tip on another thread, I looked for long folder names. I had one folder with a 36 character name. Shortening the folder name to under 32 characters seems to have fixed a repeatable crash.
  5. > Poster: NCSUCPE > Could any of your folder's names be longer than 31 characters? > If so, that might cause the quit. Yes indeed. A few subvolume scans to check the whole disk is not the problem, a find ./ -name "??????????????????????????????*" -print and one 36 character folder name in my downloads folder Intellimouse English (US and Canada) After a rename of that one folder, I can now preview the backup files for that volume without a crash. Since one of the features of Retrospect 5 running on Mac OS X is "Complete long (255 character) filenames", I am a little surprised that folders are not included, I assume that is a bug rather than a feature!
  6. I had not thought of backing up a volume at a time, I will try that too. I have narrowed by original clueless LaunchCFMApp crash down to scanning one particular volume, much like one of the other posters. I am now trying the 203 update, but the only difference so far is that I don't get a crash log, I still get the crash. I have noticed that /Library/Preferences/Retrospect/Retro.Icons (5.0) gets huge after scanning the bad volume. Scans of five "good" volumes generate a 12 KB file. Scanning one "bad" volume causes the file to grow to over 500 MB, as well as Retrospect crashing. I initially noticed the problems after running out of disk space after multiple crashes, and the file was well over a GB.
  7. The Mac OS X client is not free, it requires a Retrospect 5 client licence. To support an Mac OS X client you need Retrospect 5 Desktop/Workgroup/Server running on a Macintosh. (I am vague on the Mac OS 9 clients. I think maybe you need to upgrade the licences for Retrospect 5, but do not need to upgrade the client software? Note that the WorkGroup/Server upgrades include 20/100 licences so you may not need to worry about upgrading the client licences separately anyway.)
  8. I may have isolated my hangs to one particular client (which was previously working with 4.3 server and client), but have had a variety of other problems meaning I am not sure yet. I am installing the new server update, reinstalling on the problem client, and crossing my fingers and toes. Hopefully I will complete my first full backup with Retrospect 5 soon, and be back to smoothly running incrementals!
  9. I think adding the client by address will work, since the problem is that the client acting as a gateway and broadcasting on the default interface does not reach the server. (However, this may require Workgroup or Server rather than Desktop.)
  10. Short version: Retrospect has decided not to write further data to a tape after I got a device error. There does not seem to be a way to tell Retrospect that the tape is not really full. Is this a feature to prevent continued use of suspect media, or a bug? Details: I got a device error last night: Additional error information for device "Ecrix VXA DC" [0:3], Sense > f0 00 03 00 00 00 01 18 00 00 00 00 14 00 00 00 00 00 (ECRIX |VXA-1 |2959) Trouble writing: “3-Zeta” (4102288896), error 100 (device rejected command). and now Retrospect is asking for a new tape. The tape that got the error only has 200M instead of the typical 40G on other tapes in the same backup set. In the past after getting a hardware related error I would decide myself whether the tape was bad or the heads need cleaning or whatever, and then either continue using the tape or mark it as missing (bad). I am not sure whether I never got this particular error before, or whether Retrospect 5 behaves differently.
  11. I would suggest trying another scsi-firewire bridge if you can, it may be the problem. I am backing up on Mac OS X to a VXA via SCSI, and a VXA via firewire (first generation firewire vxa, uses the bridge supplied by Ecrix). I am having a few problems, but none that lead me to believe the tape unit is the issue.
  12. I did not have any other (visible) applications running, however it was my Mac OS 9 volume and maybe I inadvertently had Classic running. I'll experiment further. Thanks for the comments.
  13. I have been having problems with a crash in LaunchCFMApp. Initially I saw the problem three times in a row when trying to backup a particular volume on my home Mac. I would see LaunchCFMApp crash and the backup would stop. Disk Utility said the disk was fine. On my fourth try the backup ran ok. I also just had a similar crash log on my work Mac while trying to quit Retrospect after a day long stall. Anyone else seeing a LaunchCFMApp crash? Seen three times on dual 450 MHz G4 running Mac OS X 10.1.3, backing up local volumes to firewire VXA (the early model with the SCSI-firewire adaptor). Seen once on 466 MHz G4 running Mac OS X 10.1.3, backing up to SCSI VXA.
  14. Retrospect seems to sometimes stall while backing up network clients. Twice now I have come back expecting Retrospect to have filled up the first tape in a new cycle, and both times there was a Net Retry dialog on screen, and the backup had apparently stalled for over a day. Both times I was able to quit Retrospect, although it took several minutes to respond to clicking the Stop button. Anyone having similar problems, and able to shed any light? I'm going to search the Knowledge Base for the Net Retry error and attack from that end too. I may have a real network problem, although the day-long stall seems likely to be a bug too. :-) Running the backup on a G4, Mac OS X 10.1.3, SCSI VXA, 384 MB RAM, clients are over 100 Mb ethernet. I am running the Mac OS X Screen Saver, but the backup runs fine for hours after that first kicks in. The two stalls so far were while backing up Mac OS 9 clients, although they were not the first such clients to be backed up. + Normal backup using Desktop Mac (on demand) at 28/3/2002 6:27 PM [...] - 29/3/2002 11:56:48 AM: Copying Macintosh HD on Graeme Mac… 29/3/2002 11:56:48 AM: Connected to Graeme Mac 31/3/2002 4:00:46 PM: Execution stopped by operator. [Net Retry on screen, clicked stop button, long delay before Retrospect responded.] + 31/3/2002 4:11 PM: Backup Server started + Normal backup using Desktop Mac (weekend) at 31/3/2002 4:13 PM [...] - 31/3/2002 7:01:49 PM: Copying Spare Bit on MikeGill Mac… 1/4/2002 9:24:15 PM: Script execution terminated at specified stop time. 31/3/2002 7:01:49 PM: Execution incomplete. 1/4/2002 9:24:15 PM: Execution incomplete. [Net Retry on screen, clicked stop button, long delay before Retrospect responded.]
  • Create New...