Jump to content

jweisbin

Members
  • Posts

    108
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by jweisbin

  1. Not sure what you mean by paths in this context. But the bucket is called "retro-misc-ny-B2" and inside that there is a folder called the same. Inside that folder are the container folders for the individual sources. So there is a folder called "FileServer-VMFS03_Helios" which was the container from the old server. The new container is "QNAP15TB-C_Helios". It's a copy of the same folder on the new server. If I had used a client running on the source this probably would not have been an issue. But I've had too many issues with the client disappearing and having to be restarted, or saying it's busy when it's not, or "backup set reserved" whatever that means. So I rsync the data to the new server and then back that up to the cloud. So in this context I think it would have been better not to use a cloud storage group, because here I am duplicating everything onto the cloud, which is expensive.
  2. Using the "Locate" command I located the catalog file that I had copied over from the old server. In this case there was one called "retro-misc-ny-B2.rbc", and a folder called "retro-misc-ny-B2" containing files like "FileServer-macsrv001_Users_admin_Desktop.rbc", "FileServer-VMFS03_Helios.rbc", etc. Using the locate command on any of the ones in the folder does not return an error, but does not seem to do anything. Maybe I'm wrong but I don't think the "locate" command is meant to work with the actual backups - for that I would have to do a rebuild, which would take a very long time I think. (Backblaze is slow, but also relatively inexpensive as compared with Amazon for example). I'm not sure what was meant by the suggestion to "add" - in Media sets there is the big Add button which would simply add a new media set - not an existing one as far as I know. Then in the member list there is a plus button which would add another member, but that would also be a new one, not an existing one. The locate command worked insofar as I can access the full media set of backups and can restore files from it. The problem is that, from the point of view of this new server the sources are different, and therefore all new backups are going into new container folders in the cloud media set, so there is no de-duplication. But at least I have access to the older backups, which is nice. The listings indicating 0 of 4 TB space used are clearly in error, because all the backups are there and can be accessed.
  3. Sorry about the "handle" - I previously signed in using a very old account and for some reason it defaulted to that, I have no idea why. Seems to me storage groups simply don't work well with cloud storage. I started the backup and sure enough it copied everything up again as if it was all new.
  4. I have been having issues with media groups - both cloud and disk based. Retrospect thinks the media is full long before it actually is. Repair does not help. Retro version 17.0.2 on Mac OS X 10.11.6.
  5. Retrospect 15.5, on Mac Pro running OS X 10.10. Cloud backups to Amazon S3 work great, but backups to Backblaze B2 fail with many "lost connection" errors. Is there a trick to getting Backblaze to work? It's much cheaper than S3, although quite a bit slower as well.
  6. Just updated from Retro 15.1 to 15.5, on Mac Pro running OS X 10.10. Retro won't backup to a new blank, erased tape - says it's part of a backup set. Tried erasing a few times, tried restart, etc. Still no go. How do I fix this?
  7. Retro 15.1 (upgraded from 15.0) on Mac 10.11.6 Server. After updating, the server spontaneously restarts itself every few minutes. Had to go back to version 15.0, which seems stable. In addition, when upgrading the server on every version since 9, the installer says that it failed, but it apparently succeeded. And this is on multiple different systems. Can't this be fixed? I can't believe i'm the only one seeing this.
  8. Retrospect Server 14.1, running on Mac OS X 10.9.5 Server. Retrospect client 14.0.1 running on Mac OS X 10.12.6. Suddenly the server sees no clients when trying to add clients (from default interface using multicast). Existing clients are still in list, but yesterday I uninstalled and reinstalled one client. When I I tried to refresh the client, Retrospect reported "wrong client". So I deleted the client and re-added it directly via it's IP address. Now Retrospect sees only that client's boot volume, no others, even after a refresh.
  9. Yes, I did post to the wrong forum, sorry about that. Can this thread be moved?
  10. I have one client out of many for which Retrospect 14.0 running on a Mac server (OS X 10.9.5 server) keeps giving "backup client reserved" message. the only fix is to uninstall the retrospect client and reinstall it, but I've done that repeatedly and the problem returns. There is another backup client running - ChronoAgent, but I have that running on other similar machines which don't have this issue. Is there a fix?
  11. For some reason I was under the impression that you could backup to multiple backup sets at same time. But I can't find anything in the documentation. I thought I would set up a disk backup set and a tape backup set that would be perfectly in sync. I set a script to backup to those two media sets, but when I run the script it asks me which media set to back up to. What am I missing? Do I first have to do a copy media set from the disk to the tape?
  12. Uninstall and reinstall worked. Apparently, the latest uninstaller deletes retroclient.state on its own.
  13. Recently updated from 13.0.1 to 13.5. I have one client for his mounted volumes used to show up but now appear with the missing "~" mark. I tried to refresh, no change. Then I deleted to source and set it up again, and now Retrospect only sees the boot volume. How to fix?
  14. yea, tried all that, tapes won't come out. I'm just suspicious that not one but two drives failed within a week of each other and right after upgrading.
  15. I have one LTO-5 drive (HP Ultrium) and two LTO-4 tape drives (HP Ultrium 1760 SAS). Shortly after upgrading to Retrospect 11 both LTO-4's refuse to eject the last cartridge used. I have a suspicion that Retrospect 11 somehow caused this. Anyone else seen this happen?
  16. this is inside the office, client and server are both behind the firewall. Besides, it can back up from the client, so why wouldn't be able to write to the client? Also, this problem only surfaced in version 11.
  17. Running Retro 11.0.1 on a OS X 10.6.8 server, client is another server running 11.0.1 client. Trying to restore to a folder on the client server. Had no problems in Retro 10.5, now getting message that the destination is "locked or not writeable". Checked permissions and they are wide open. server is Mac Pro quad core running 10.6.8, ATTO H680 card with HP LTO-5 tape (but restoring from a disk media set) client is a mac mini running 10.9.2 server. Destination is on an attached thunderbolt array made by Areca also getting this on one client so far: Scanning incomplete, error -1010 ( API request bad)
  18. and another client error: "Scanning incomplete, error -1010 ( API request bad)"
  19. Running Retro 11.0.1 on a OS X 10.6.8 server, client is another server running 11.0.1 client. Trying to restore to a folder on the client server. Had no problems in Retro 10.5, now getting message that the destination is "locked or not writeable". Checked permissions and they are wide open. server is Mac Pro quad core running 10.6.8, ATTO H680 card with HP LTO-5 tape (but restoring from a disk media set) client is a mac mini running 10.9.2 server. Destination is on an attached thunderbolt array made by Areca
  20. Retro 10.1, Mac OS X 10.6.8, backup set is on a SATA attached drive. Ran the Restore wizard, "select files or folders", selected a backup set to restore, selected a destination (on another volume), Retro says "needs media...". When I click the button to locate the media and click "add" (it's saying it can't find backup set 1-BigBak, so I point it to that folder), nothing happens, it still won't restore. I have gone through these exact steps on another server and not had this problem, so it seems like another random bug.
  21. Retrospect 10.1 is basically unusable with clients. They randomly disappear, and the server has to be restarted, or sometimes they have to be deleted and re-added back as sources. In addition, when these errors occur, retro is not send notification emails, even though the email test function works and I get those emails. It also randomly says "media set is in use" when it isn't. The only fix for that is to repair the media set, apparently. I have these problems across three servers and many clients so it is not a random bug. How do people live with this junk? I didn't buy the support package so no doubt they will charge me for support. This is shameful.
  22. clarification: the server is 10.1, the client is 10.0.1 My mistake - I thought that the "bug reports" section would be looked at by Retrospect.
  23. Mac OS X 10.6.8 server, Mac Pro 8 core 2.8 ghz, 16 GB RAM, ATTO H680 SAS card, HP LTO-5 tape drive, Retro 10.0.1 server and clients Having same problems on multiple servers: Retro randomly loses communication with clients every few days, and does NOT report that error via email Retro says "media set is in use" when it isn't - have to repair the media set (a disk) and then it runs again - again no error email. Seems to be better if I leave the console running on the server itself. RetroISA causing kernel panic on clients and must be disabled. I have reported this previously and gotten NO replies.
×
×
  • Create New...