Having trouble connecting with AFP

[loadbang]

Trying to use Time Machine feature. However, I keep on having connection issues.

Time Machine on my Macs can see the backup drive in System Preferences, but is stuck at "Looking for Backup Disk" when trying to create a backup.

Tried finder, it shows as an AFP volume in the sidebar, but when clicking I get "“RT-AC68U-A1C8.local” can’t be opened because the original item can’t be found." error message.

 

[mattiL]

I have similiar problem on my Macbook Pro, after it has been in sleep mode WITH power connected it won't find my Synology NAS.

If I put it to sleep WITHOUT power connected and wake it up, it works!

 

[loadbang]

What I have found is a process cnid_dbd is running at 100% on the AC68. Hmm.

Problem is, I can't kill the process or else it corrupts the drive.

 

[loadbang]

My system.log is constantly:

Sep 29 11:49:57 afpd[5577]: read: Connection reset by peer
Sep 29 11:49:57 cnid_metad[5645]: Multiple attempts to start CNID db daemon for "/tmp/mnt/seagate3tb" failed, wiping the slate clean...
Sep 29 11:49:57 cnid_dbd[5645]: Couldn't open extension maping file /usr/etc/extmap.conf
Sep 29 11:49:57 cnid_dbd[5645]: Couldn't load extension -> type/creator mappings file "/usr/etc/extmap.conf"
Sep 29 11:49:57 cnid_dbd[5645]: error deleting key/value from cnid2.db: DB_RUNRECOVERY: Fatal error, run database recovery
Sep 29 11:49:57 cnid_dbd[5645]: dbd_delete: Unable to delete entry for CNID 5931
Sep 29 11:49:57 cnid_dbd[5645]: error aborting transaction: DB_RUNRECOVERY: Fatal error, run database recovery
Sep 29 11:49:57 cnid_dbd[5645]: error closing database name.db: DB_RUNRECOVERY: Fatal error, run database recovery
Sep 29 11:49:57 cnid_dbd[5645]: error closing database didname.db: DB_RUNRECOVERY: Fatal error, run database recovery
Sep 29 11:49:57 cnid_dbd[5645]: error closing database devino.db: DB_RUNRECOVERY: Fatal error, run database recovery
Sep 29 11:49:57 cnid_dbd[5645]: error closing database cnid2.db: DB_RUNRECOVERY: Fatal error, run database recovery
Sep 29 11:49:57 cnid_dbd[5645]: error closing DB environment: DB_RUNRECOVERY: Fatal error, run database recovery

 

[loadbang]

So, I touched /tmp/netatalk/extmap.conf and all seems ok so far. The conf file was missing.

 

[loadbang]

Only problem is, when you reboot this file vanishes.

Checked "afpd -V"

# afpd -V
afpd 3.0.5 - Apple Filing Protocol (AFP) daemon of Netatalk

This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; either version 2 of the License, or (at your option) any later
version. Please see the file COPYING for further information and details.

afpd has been compiled with support for these features:

  AFP versions:   2.2 3.0 3.1 3.2 3.3
  CNID backends:   dbd last tdb
  Zeroconf support:   No
  TCP wrappers support:   No
  Quota support:   No
  Admin group support:   Yes
  Valid shell checks:   Yes
  cracklib support:   No
  EA support:   ad | sys
  ACL support:   No
  LDAP support:   No
  D-Bus support:   No
  DTrace probes:   No

  afp.conf:   /usr/etc/afp.conf
  extmap.conf:   /usr/etc/extmap.conf
  state directory:   /tmp/netatalk/
  afp_signature.conf:   /tmp/netatalk/afp_signature.conf
  afp_voluuid.conf:   /tmp/netatalk/afp_voluuid.conf
  UAM search path:   /usr/lib/netatalk/
  Server messages path:   /root/asuswrt-merlin/release/src-rt-6.x.4708/router/arm-uclibc/install/netatalk-3.0.5/usr/var/netatalk/msg/

extmap.conf should be in /usr/etc/extmap.conf but doesn't exit in firmware, and I cannot write to /usr/etc

Help!


Would be advantageous if AFP was updated to a newer version. Merlin crowd can we have this compiled in to later versions?

I'm also wondering, with afp_voluuid.conf being in the /tmp directory, this is also causing a headache as it is recreated every boot, meaning my Time Machine backups wont backup again as the UUID changes. Time Machine is effectively broken. Would like someone with more knowledge to triage this before filing as a bug.

 

[Raiu]

so this is still happening, I just noticed my time machine has stopped working as well..... So this has been going on since Sept? wow.

 

[mattiL]

I got my TM backup working by making two changes in System Preferences > Energy Saver, I unchecked Wake for WiFi and Enable Power Nap and now backup works again!

 

[Jerry12]

Only problem is, when you reboot this file vanishes.
...

I'm also wondering, with afp_voluuid.conf being in the /tmp directory, this is also causing a headache as it is recreated every boot, meaning my Time Machine backups wont backup again as the UUID changes. Time Machine is effectively broken. Would like someone with more knowledge to triage this before filing as a bug.

Sometimes I have to manually reconnect Time Machine to the server. In OSX Preferences > Time Machine, clicking Select Disk... and hovering over the disk shows that it's connected to afp://admin@RT-AC3200-1234._afpovertcp._tcp.local./Backups.backupdb and not to the similar-looking entry with the hover text afp://RT-AC3200-1234.local/Backups.backupdb . This seems to be needed.

With AsusWRT-Merlin, you can write a user script that runs before the AFP and TM servers start, modifies their config files, and touches /tmp/netatalk/extmap.conf (or stops the servers, modifies their config files, and restarts them). I have not tried this, so if the docs are unclear, search and ask on the forums.

Also see http://www.snbforums.com/threads/how-to-set-up-rt-timemachine-server.25368/ and please add discoveries there on how to make TM work.