[Box Backup] error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry

Johann Glaser boxbackup@fluffy.co.uk
Fri, 07 Sep 2007 10:35:42 +0200


Hi!

Since a few days we have a problem with the BoxBackup storage. The
daemon.log says the following (except housekeeping)

Sep  7 03:48:19 clara bbackupd[23240]: Incoming connection from local (UNIX socket)
Sep  7 03:48:19 clara bbackupd[23240]: Connection from command socket
Sep  7 03:48:19 clara bbackupd[23240]: Beginning scan of local files
Sep  7 03:48:19 clara bbackupd[23240]: Opening connection to server clara...
Sep  7 03:48:19 clara bbstored[23252]: Incoming connection from 128.131.80.105 port 56481 (handling in child 21264)
Sep  7 03:48:20 clara bbstored[21264]: Certificate CN: BACKUP-1 
Sep  7 03:48:20 clara bbstored[21264]: Login: Client ID 00000001, Read/Write
Sep  7 03:48:20 clara bbackupd[23240]: Connection made, login successful
Sep  7 04:09:06 clara bbstored[21264]: Connection statistics for BACKUP-1: IN=747553 OUT=660125 TOTAL=1407678 
Sep  7 04:09:06 clara bbstored[21264]: in server child, exception Connection Protocol_Timeout (Probably a network issue between client and server.) (7/41) -- terminating child
Sep  7 04:11:50 clara bbstored/hk[23253]: different (store,calc): acc 0x00000001, used (28649525,28649863), old (19073731,19074043), deleted (102167,102167), dirs (16255,16256)
Sep  7 04:32:45 clara bbackupd[23240]: SSL err during Write: error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry
Sep  7 04:32:45 clara bbackupd[23240]: Exception caught (Connection TLSWriteFailed (Probably a network issue between client and server.) 7/33), reset state and waiting to retry...
Sep  7 04:32:55 clara bbackupd[23240]: File statistics: total file size uploaded 549290554, bytes already on server 187528808, encoded size 743829

Interrestingly, the server complains at 4:09 about a timeout, and the
client complains at 4:32 about an SSL error. I started the backup
several times yesterday, and there was always a difference of approx. 25
minutes.

I checked the backup store and all files have the same owner and
permissions (except ./write.lock which has 600).

A few days ago I increased the MaximumDiffingTime to 20000, but this
worked for 2 days without a problem.

I've restarted both, the client and the server process, and remounted
the slow NFS storage. Nothing helped.

Can you please help me to find out what is wrong?

Thanks
  Hansi

-- 
Johann Glaser                          <glaser@ict.tuwien.ac.at>
             Institute of Computer Technology, E384
Vienna University of Technology, Gusshausstr. 27-29, A-1040 Wien
Phone: ++43/1/58801-38444                Fax: ++43/1/58801-38499