Jump to content


  • Content count

  • Joined

  • Last visited

Community Reputation

0 Neutral

About jdpal

  • Rank
    Occasional forum poster
  1. Price out the upgrade to 7.6 and price out how much they've paid you so far to make it work with 7.0.. =)
  2. Interesting! I didn't know that. I guess that explains why it works.
  3. Hi Mayoff, Regarding the resource forks, I did a test. I took a folder full of Mac fonts which are basically all resource forks, the data fork is empty. I then backed that up via Retrospect (client backup) and restored it to a local NTFS formatted disk. As expected, the files all restored showing size 0k (in Windows), but when viewed over the network from a Mac (the server is Windows 2003 R2 with the file sharing for mac services installed), the font files appear as perfectly intact Mac font files. It appears that Retrospect is in fact transferring the resource forks. Is that correct?
  4. jdpal

    Very slow verify

    Hi Mayoff, We did a firmware upgrade which solved the performance problem with reads. Thanks!
  5. jdpal

    Very slow verify

    Hi Mayoff, I ran a cleaning tape through the drive twice. The drive did need to be cleaned (and apparently there was not a cleaning tape in the autoloader for quite some time). It is a Sony AIT-4 DC drive in a Qualstar robotic library, attached via SCSI. I performed the following tests: I took a brand new tape and created a new tape set for it, adding it as the only member. I then backed up a 10GB folder full of MP3 files. The backup averaged 1266MB/Min. I had it set to media verify. The media verify chugged along at about 65MB/Min before I canceled it. Then I did a restore from that new tape to a hard drive array directly attached to the retrospect server. The restore also chugged along at about 65MB/Min. I did not switch SCSI cables simply because the writing is fine but the reading isn't. What do you suggest? Is there a way in Retrospect to get more information about the slowdown? Thanks!
  6. Hi, I am running Retro 7.6 on W2k3. I recently expanded the C: partition from 12GB to 50GB as I heard it would help media verify operation. It appears that my disk sets do indeed do media verifies faster. However, I have a tape set that crawls on verify no matter if it's media verification or thorough verification. Here is the log from the latest job, which I cancelled after 2 days... + Recycle backup using Servers at 11/21/2008 8:00 PM (Execution unit 2) To Backup Set Servers B... 11/21/2008 8:00:01 PM: Recycle backup: The Backup Set was reset - 11/21/2008 8:00:01 PM: Copying Accounting on CLEESE 11/21/2008 8:06:21 PM: Snapshot stored, 91 KB 11/21/2008 8:06:22 PM: Comparing Accounting on CLEESE 11/21/2008 8:40:58 PM: Execution completed successfully Completed: 149 files, 538.7 MB Performance: 30.4 MB/minute (371.4 copy, 15.8 compare) Duration: 00:40:57 (00:05:33 idle/loading/preparing) - 11/21/2008 8:41:01 PM: Copying Repositories on Dev Server 11/21/2008 8:45:23 PM: Snapshot stored, 2,085 KB 11/21/2008 8:45:28 PM: Comparing Repositories on Dev Server 11/21/2008 11:56:53 PM: Execution completed successfully Completed: 8854 files, 4.0 GB Performance: 42.0 MB/minute (1054.5 copy, 21.4 compare) Duration: 03:15:52 (00:01:59 idle/loading/preparing) - 11/21/2008 11:56:54 PM: Copying ebooksearch on FERRELL 11/21/2008 11:57:44 PM: Snapshot stored, 17 KB 11/21/2008 11:57:46 PM: Comparing ebooksearch on FERRELL 11/21/2008 11:58:27 PM: Execution completed successfully Completed: 21 files, 755 KB Performance: 9.8 MB/minute (6.3 copy, 44.2 compare) Duration: 00:01:33 (00:01:24 idle/loading/preparing) - 11/21/2008 11:58:29 PM: Copying vPage on FERRELL 11/22/2008 12:07:34 AM: Snapshot stored, 20.9 MB 11/22/2008 12:07:54 AM: Comparing vPage on FERRELL 11/22/2008 2:03:39 AM: Execution completed successfully Completed: 138274 files, 2.9 GB Performance: 47.9 MB/minute (392.1 copy, 25.5 compare) Duration: 02:05:09 (00:03:03 idle/loading/preparing) - 11/22/2008 2:03:41 AM: Copying Tomcat 6.0 on MAHER 11/22/2008 2:04:38 AM: Snapshot stored, 86 KB 11/22/2008 2:04:39 AM: Comparing Tomcat 6.0 on MAHER 11/22/2008 2:04:56 AM: Execution completed successfully Completed: 246 files, 47.7 MB Performance: 439.5 MB/minute (357.0 copy, 714.1 compare) Duration: 00:01:15 (00:01:01 idle/loading/preparing) - 11/22/2008 2:04:59 AM: Copying Tomcat 6.0 on lewis 11/22/2008 2:10:49 AM: Snapshot stored, 95 KB 11/22/2008 2:10:51 AM: Comparing Tomcat 6.0 on lewis 11/22/2008 2:55:11 AM: Execution completed successfully Completed: 309 files, 4.2 GB Performance: 183.8 MB/minute (867.0 copy, 102.8 compare) Duration: 00:50:12 (00:03:50 idle/loading/preparing) - 11/22/2008 2:55:13 AM: Copying syncdata on lewis 11/22/2008 6:43:45 AM: Snapshot stored, 52.4 MB 11/22/2008 6:44:34 AM: Comparing syncdata on lewis 11/24/2008 9:28:33 AM: Execution stopped by operator Remaining: 141244 files, 152.9 GB Completed: 71156 files, 124.4 GB Performance: 126.2 MB/minute (1280.7 copy, 41.9 compare) Duration: 2 06:33:21 (00:15:59 idle/loading/preparing) 11/24/2008 9:29:09 AM: Execution stopped by operator Total performance: 118.9 MB/minute Total duration: 2 13:28:33 (00:32:51 idle/loading/preparing) Any suggestions for what to look at to increase the verify speeds? This is on thorough verification but the results have been similar with media verify. Thanks!
  7. Unfortunately that's what I did - just a backup of that mailbox. It generated a 100mb log :taptap:
  8. Hi Mayoff, I set the trees and volumes debug to 6, and the backup job generated a nice 100MB log. Unfortuantely Retrospect was not able to export the entire log, it gave up after about 740,000 lines. That cut off the beginning of the log which may be where the useful info is. Is there any reason why the errors button is greyed out?
  9. Hi Mayoff, I changed the mailbox backup logging level to 6 and restarted Retrospect. I then ran the mailboxes script again. No further information is reported in the log and I still get the same exact errors as before. I still cannot click on the "errors" button in the job history log. Any other suggestions?
  10. Hi Mayoff, Thanks for the quick response. Unfortunately the errors button is greyed out for everything in my history log. This is Retrospect 7.6.111
  11. Hi, When I back up some users' mailboxes, I get errors like this: - 10/31/2008 10:44:43 PM: Copying User1 on Mailserver T-14: MBDataInputStream::PrepareLongProperty: Failure (winerr = 0x8004011b) to read long property 0x1009. File "User\Deleted Items\0000000041777b9ace11e2449d6126c7f83fd2d20700ba5af3afafc8934ea61d1d91775c25de00002b034f7c0000ba5af3afafc8934ea61d1d91775c25de000034b5e2b30000": can't read, error -1012 (feature unsupported) 10/31/2008 10:45:12 PM: Snapshot stored, 3,465 KB 10/31/2008 10:46:38 PM: 1 execution errors Remaining: 1 files, 18 KB Completed: 45 files, 1,174 KB Performance: 0.7 MB/minute Duration: 00:01:55 (00:00:26 idle/loading/preparing) - 10/31/2008 10:46:40 PM: Copying User2 on Mailserver File "User2\Calendar\000000001fbf487112efd946bb973eb441dd58450700ba5af3afafc8934ea61d1d91775c25de00002b02c3d1000026a307ad8fe4b84794c046b9a5c148de0000349ad6070000": can't read, error -1101 (file/directory not found) 10/31/2008 10:48:01 PM: Snapshot stored, 21.1 MB 10/31/2008 10:50:21 PM: Execution completed successfully Remaining: 1 files, 24 KB Completed: 149 files, 10.1 MB Performance: 4.5 MB/minute Duration: 00:03:40 (00:01:26 idle/loading/preparing) I think these are bad/corrupted items within the users mailbox. I was wondering if anyone knows how to identify which items these are (based on the data given in the error)? From there I could ask the user to delete the item. Thanks!
  12. Is there any way to do this process manually? We are not running Retrospect on the Exchange server - it is a client backup. Thanks!
  13. Hi Lars, If the PST is copied to the server automatically, and then backed up, that should be fine, but be careful because Microsoft strongly advises against storing PSTs on a network volume while they are in use.
  14. Hi, I am running Retrospect 7.6 multi-server on W2k3 server standard edition. I am backing up several Mac file servers using the Retrospect client that was the latest version (until a few days ago when a new version came out). Some of the files/folders/paths, while legal on the Mac filesystem (HFS+ Journaled) are not okay under Windows. I have seen a folder path refuse to restore (to an NTFS volume attached to the backup server, not to the original source) because the path is too long. There are also some illegal characters in various files and folders. I understand that the above is a limitation of Windows and that is fine. My question is - in terms of Mac-legal characters in file names: What will Retrospect do with these filenames if I ask it to restore them to an NTFS volume? Will it refuse to restore? Will it strip them/substitute them? Thanks!