Disconnection during long transfers?

[Cubytus]

Hi there,

just noticed an annoying behavior. During a multi-gigabyte transfer, it happened more than once that the NAS ejects itself, but stays available when attempting connection.

What is strange is, the transfer isn't interrupted, but keeps going and succeeds.

I first thought about the overheat protection kicking in, but the drive stays on, and the temperature reading stays below the treshold.

How to find what's causing these disconnections?

[Jocko]

Hi Cubytus,

I assume that you connect to nas by an afp access.

So you can read its log file (/var/log/afpd.log) to search if there are some reported errors.

[Cubytus]

Unfortunately, there's no such log file.

Code: Select all

/var/log # ls
cups  httpd  lastlog  messages  samba  wtmp


[Jocko]

Indeed you do not have a log file as log level is too low (warn).

So edit the file /usr/etc/afp.conf and change

Code: Select all

log level = default:warm

by setting a higher level instead of warn.
value can be: note, info, debug, debug6, debug7, debug8, debug9, maxdebug.

I believe with info or debug level you should have information.

Then restart afp daemon

Code: Select all

killall afpd
/usr/sbin/afpd

[Cubytus]

Ok, I am now attempting to move a large, 15GB file to fvdw, then I will try from fvdw…

So far it seemed to be working, at a decent 45MiB/s. Any way to get even more speed from this network HDD?

Since the bug didn't reproduce, I don't know what would have gone wrong.

[Jocko]

Hi Cubytus,

I forgot to say two things:
- At each time, you start afpd server from the web-interface (or change settings on share or user account) or reboot the nas, conf file will be overwritten. So you need to edit it again to set an higher log level.
-read/write speed via afp server can be strongly degraded with this log level (for example with maxdedug level)