2:19 p.m.
I've started getting an error message I don't understand when trying to
sync between two peers:
jwd@maple ~ $ syncevolution --sync refresh-from-server redwood-server todo
[INFO] pat: inactive
[INFO] personal-projects: inactive
[INFO] work-calendar: inactive
[INFO] personal-blocked: inactive
[INFO] events: inactive
[INFO] memo: inactive
[INFO] work-blocked: inactive
[INFO] addressbook: inactive
[INFO] commute: inactive
[INFO] calendar: inactive
[INFO] work: inactive
[INFO] SoupTransport Failure: http://redwood:9000/syncevolution via libsoup: D-Bus
server done
^X[INFO] SyncContext: resend previous request #1
[INFO] SoupTransport Failure: http://redwood:9000/syncevolution via libsoup: D-Bus
server done
^[[1;5D^[[1;5D[INFO] SyncContext: resend previous request #1
[INFO] SoupTransport Failure: http://redwood:9000/syncevolution via libsoup: D-Bus
server done
The SoupTransport error messages repeat until SE gives up. The same
thing happens no matter which source (or all) I try to sync. The same
thing also occurs if I try a --sync refresh-from-server.
This is an SE 1.1 client talking to an SE 1.1 server, which is using the
Evolution backends. The log on the server side shows _two_ _successful_
sessions for each "SoupTransport Failure" message on the client side.
These two peers were syncing with each other yesterday. The change I
mad this morning was to add a source to the configuration on both sides.
However, that source is not involved in the above attempt.
Can anyone suggest where to look to troubleshoot this problem?
-- John
6:46 p.m.
On Do, 2011-01-06 at 03:19 +0000, John Dykstra wrote:
I've started getting an error message I don't understand when
trying to
sync between two peers:
jwd@maple ~ $ syncevolution --sync refresh-from-server redwood-server todo
[INFO] pat: inactive
[INFO] personal-projects: inactive
[INFO] work-calendar: inactive
[INFO] personal-blocked: inactive
[INFO] events: inactive
[INFO] memo: inactive
[INFO] work-blocked: inactive
[INFO] addressbook: inactive
[INFO] commute: inactive
[INFO] calendar: inactive
[INFO] work: inactive
[INFO] SoupTransport Failure: http://redwood:9000/syncevolution via libsoup:
D-Bus server done
^X[INFO] SyncContext: resend previous request #1
[INFO] SoupTransport Failure: http://redwood:9000/syncevolution via libsoup:
D-Bus server done
^[[1;5D^[[1;5D[INFO] SyncContext: resend previous request #1
[INFO] SoupTransport Failure: http://redwood:9000/syncevolution via libsoup:
D-Bus server done
The SoupTransport error messages repeat until SE gives up. The same
thing happens no matter which source (or all) I try to sync. The same
thing also occurs if I try a --sync refresh-from-server.
This is an SE 1.1 client talking to an SE 1.1 server, which is using the
Evolution backends. The log on the server side shows _two_ _successful_
sessions for each "SoupTransport Failure" message on the client side.
These two peers were syncing with each other yesterday. The change I
mad this morning was to add a source to the configuration on both sides.
However, that source is not involved in the above attempt.
Can anyone suggest where to look to troubleshoot this problem?
1. use the improved syncevo-http-server.py script from master on
the server (can be combined with the existing 1.1.1 (!)
installation - please update from 1.1 to 1.1.1):
http://meego.gitorious.com/meego-middleware/syncevolution/blobs/raw/maste...
2. run it with -d to get the full output
3. check for anything suspicious, or send it to me (directly,
probably too large for the list)
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
9:24 p.m.
New subject: [SyncEvolution] resolved: smltk DevInf buffer size (was: Re: SoupTransport Failure: http://../syncevolution via libsoup: D-Bus server done)
Hello!
John and I kept analyzing this offline. Thanks to John's help, I found
the root cause: he has so many sources configured, that generating the
DevInf for them exceeds a hard-coded limit of 12000 bytes in the SyncML
Toolkit library.
This in turn causes libsynthesis to log "SmlException: SyncML Toolkit
error=0x2002, smlResultsCmd, smlerr=8194" in the server HTML log file
before it bails out without sending anything back to the client (because
the message couldn't be encoded), which then leads to the "D-Bus server
done" error (because that's all that the HTTP server sees).
Unfortunately there is no nice solution short of patching libsynthesis
and the SyncML Toolkit that it contains. Is that important enough for a
1.1.2 release or can it wait until 1.2?
John, the patch is attached.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
4:27 a.m.
New subject: [SyncEvolution] resolved: smltk DevInf buffer size (was: Re: SoupTransport Failure: http://../syncevolution via libsoup: D-Bus server done)
On Mon, 2011-01-17 at 11:24 +0100, Patrick Ohly wrote:
John, the patch is attached.
Unfortunately, the attached diff is zero-length.
-- John
5:57 a.m.
New subject: [SyncEvolution] resolved: smltk DevInf buffer size (was: Re: SoupTransport Failure: http://../syncevolution via libsoup: D-Bus server done)
[Resending, initial mail didn't make it through.]
Hello!
John and I kept analyzing this offline. Thanks to John's help, I found
the root cause: he has so many sources configured, that generating the
DevInf for them exceeds a hard-coded limit of 12000 bytes in the SyncML
Toolkit library.
This in turn causes libsynthesis to log "SmlException: SyncML Toolkit
error=0x2002, smlResultsCmd, smlerr=8194" in the server HTML log file
before it bails out without sending anything back to the client (because
the message couldn't be encoded), which then leads to the "D-Bus server
done" error (because that's all that the HTTP server sees).
Unfortunately there is no nice solution short of patching libsynthesis
and the SyncML Toolkit that it contains. Is that important enough for a
1.1.2 release or can it wait until 1.2?
John, the patch is attached. [And here is a link:]
http://meego.gitorious.com/meego-middleware/libsynthesis/commit/722cb0154...
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
4262
days inactive
4273
days old
syncevolution@synevolution.org
4 comments
3 participants
participants (3)
-
John Dykstra -
John Dykstra -
Patrick Ohly