AFPD TimeMachine Error After Upgrade To 6.6.6


Recommended Posts

Hello, i recently upgraded to 6.6.7RC2 and AFP connection wouldnt work. I then downgraded to 6.6.6 and my AFP drives mount in macOS and i can transfer files through finder, but TimeMachine no longer works. I even tried going back to 6.6.5 and my TimeMachines still doesn't work.

 

I tried restarting server, restarting client mac, and even stopping AFP and SMB in unraid settings, and re-starting them. Still the error persist. What else can i try? I checked to make sure the shares are ok, and that they have AFP enabled.

 

The TimeMachine connects to a disk share, that is setup as AFP with secure and my user account is listed as read/write.

 

Here is the errors im getting.

 

2079195788_unraidafpderror.thumb.png.ed12b7a38949c7dfa6433fa58f2df574.png

 

Link to comment

Post your diagnostics zip.

 

You can try rebuilding the Netatalk database, as follows. Unmount all AFP-mounted shares (or just shut down your Mac). Then from the Unraid server's command line, or using MC or Krusader or Dolphin dockers, delete the .AppleDB folder in the root of each user share. Then start your Mac or re-mount the shares. They will take a while longer to mount than usual because the database has to be rebuilt. You can improve speed and reliability by moving the databases off of the array and onto your cache pool. I've posted information about how to do this before. I'll try to find a link to it.

 

Edit: Here, 5th paragraph for the .AppleDB stuff. The whole thing for information about Time Machine.

 

Edited by John_M
Added link
Link to comment

Thanks John_M i was able to re-upgrade to 6.7RC2 and found a setting to turn on timemachine in the SMB settings, then i was able to setup the disk share as TimeMachine under SMB. My mac can now perform TimeMachine Backups. I do have the dB path for all my shares on the cache, drive in a cache only share. The one i forgot to do one for is my disk2 share. ill try that tonight see if it helps. I might stick to SMB if it continues to work tho.

 

Link to comment
  • 4 weeks later...

Hi All,

since years, we are using a Synology system for backing up all our Mac computers. The backups runs over the AFP-Protocol without any issues.

Because of power (and money) saving, we want to change to a unRAID system for these backus.

So we have bought a Supermicro storage server SSG-6029P-E1CR12L (X11DPH-T mainboard), 96GB RAM, Intel Xeon 4110 CPU, 10 of 14TB WD Ultrastar DC HC530 HDDs, 2 Intel SSDSC2KB019T8 1.9TB SATA-SSDs, 2 Intel SSDPELKX020T8 2.0TB NVME-SSDs. On this Hardware, unRAID 6.6.6 was installed and updated to 6.6.7. Since weeks, we don't get one simple TimeMachine backup from a brand new Mac which has about 270GB to backup (over AFP). We are getting a lot of errors like described above. Only one backup was finished successfully, but at the next backup TimeMachine displayed that the backup must be checked and after that check, TimeMachine displayed that a new backup has to be created (the previous backup was automatically renamed to ...purgeable.sparsebundle). Also the backup time is very slow (24+ hours). The Mac and unRAID is connected with a 10GBit/sec. link to a 10GBit switch. Have also tried many settings without success.

 

My question here is, is there a best practice documentation on how to setup unRAID properly for running Time Machine backups, like we have done with our old Synology system? Or maybe unRAID isn't the right solution for their requirement.

 

Are there any ideas ... ?

Thanks in advance. Reini.

 

P.S.: Btw. the rest of the unRAID services runs well.

Edited by Reini
Link to comment
1 hour ago, Reini said:

Are there any ideas ... ?

Did you read the earlier posts in this thread and follow the advice I gave to alfredo_2020? The link I gave has step by step instructions for setting up Time Machine over AFP and explains why sparse bundle disk images are so fragile and how to minimise the risk of corruption.

 

 

If Time Machine detects corruption to the sparse image it creates a new one from scratch. The purgeable sparse image gets deleted during the cleanup phase.

 

If you don't want to use AFP and your Macs are all running Mojave or High Sierra then consider updating Unraid to 6.7.0-rc and using SMB instead.

 

Link to comment
On 3/6/2019 at 1:23 PM, John_M said:

Did you read the earlier posts in this thread and follow the advice I gave to alfredo_2020?

No, but thanks for the advice. I've done almost all the points except for moving the AppleDB into the SSD cache pool. Will try this and see what's happen ...

Link to comment
Quote

Did you read the earlier posts in this thread and follow the advice I gave to alfredo_2020?

I have done everything exactly as described but no success.

One Mac has done the TimeMachine Backup 5 times without issues. At the 6th time, TimeMachine said, a new backup is needed and failed. From that time, I have seen many errors/failures at the unRAIDs syslog (see below). Also from that time, no Mac is able to do a TimeMachine backup anymore and skip the backups to unRAID. At the same time, all Time Machine backups to the old Synology system still works without any issues. After rebooting unRAID the afpd errors disappears and Time Machine is able to doing backups again, but only until the afpd error comes back and the loop starts from the beginning.

It seams the the AFP implementation isn't stable enough!?

 

Any other ideas?

 

Quote

Mar 13 09:01:30 unRAID afpd[19182]: INTERNAL ERROR: Signal 11 in pid 19182 (3.1.11)
Mar 13 09:01:30 unRAID afpd[19182]: ===============================================================
Mar 13 09:01:30 unRAID afpd[19182]: PANIC: internal error
Mar 13 09:01:30 unRAID afpd[19182]: BACKTRACE: 10 stack frames:
Mar 13 09:01:30 unRAID afpd[19182]: #0 /usr/lib64/libatalk.so.18(netatalk_panic+0x22) [0x14d57b449c92]
Mar 13 09:01:30 unRAID afpd[19182]: #1 /usr/lib64/libatalk.so.18(+0x40dbc) [0x14d57b449dbc]
Mar 13 09:01:30 unRAID afpd[19182]: #2 /lib64/libc.so.6(+0x407d0) [0x14d57a6927d0]
Mar 13 09:01:30 unRAID afpd[19182]: #3 /usr/lib64/libatalk.so.18(bformat+0x109) [0x14d57b42c159]
Mar 13 09:01:30 unRAID afpd[19182]: #4 /usr/sbin/afpd(of_closefork+0x5e) [0x42924e]
Mar 13 09:01:30 unRAID afpd[19182]: #5 /usr/sbin/afpd(afp_closefork+0x83) [0x427003]
Mar 13 09:01:30 unRAID afpd[19182]: #6 /usr/sbin/afpd(afp_over_dsi+0x575) [0x40cc95]
Mar 13 09:01:30 unRAID afpd[19182]: #7 /usr/sbin/afpd(main+0xa55) [0x40ac05]
Mar 13 09:01:30 unRAID afpd[19182]: #8 /lib64/libc.so.6(__libc_start_main+0xeb) [0x14d57a6760ab]
Mar 13 09:01:30 unRAID afpd[19182]: #9 /usr/sbin/afpd(_start+0x2a) [0x40ae9a]
Mar 13 09:01:36 unRAID afpd[24727]: afp_disconnect: primary reconnect failed
Mar 13 09:01:36 unRAID afpd[24727]: afp_closefork: of_find(38) could not locate fork
Mar 13 09:01:36 unRAID afpd[24727]: get_tm_bandsize("/mnt/user/TimeMachine/MacPro.purgeable.sparsebundle/Info.plist"): No such file or directory

 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.