Archived

This forum has been archived. Please start a new discussion on GitHub.

inaccurate timeout in ice call

suds
suds
in Help Center
Hi there:
I rencently ecnounter a wierd thing when using ice calll from client to server.
Say , I set Ice.Override.Timeout to 30000 when client invoke server through ice . and at that time ,server is busy and temporarily blocked for a while. but only after 6000ms ,client raise a timeout exception.
And this problem is not easy to reproduce and I don't know clearly which condition can cause this problem.
Is there any one else who encounter this problem ? and is there any solution or suggestion for this issue ?
Any help or suggestion is welcome.
Thank you very much!

Suds

Comments

  • benoit
    benoit Rennes, France
    Hi,

    By default, the Ice runtime retries the invocation if at-most-once semantics can still be guaranteed or if the invocation is idempotent. Most likely, this is what is happening here.

    You can enable retry/network tracing with Ice.Trace.Network=2 and Ice.Trace.Retry=1 to trace when the runtime retries and establishes network connections. You can disable retries by setting Ice.RetryIntervals=-1.

    Cheers,
    Benoit.
  • benoit
    benoit Rennes, France
    Actually, I might have misread your post. If the timeout kicks in after 6000 ms , it's most likely another issue and I'm not sure what could be the reason (setting Ice.Trace.Network=3 might give us more clues). Or is this a typo and you meant to write 60000 ms?

    In any case, for more information on connection timeouts, I recommend to check out the manual here.

    Cheers,
    Benoit.

Categories