Opened 11 years ago

Closed 11 years ago

#3307 closed defect (fixed)

Higer ecm time in a client server configuration

Reported by: apprendista Owned by:
Priority: critical Component: ! Please select...
Severity: medium Keywords:
Cc: Sensitive: no

Description

8638
ECM time is higher if release 8638 is used on the client

alwais

looking at the log of the client

Attachments (1)

tryfix1.patch (2.7 KB ) - added by theparasol 11 years ago.
No wait on fd in error, wait max 3000ms for fd to become ready

Download all attachments as: .zip

Change History (9)

comment:1 by apprendista, 11 years ago

Priority: Please fill incritical
Severity: Please fill inmedium

comment:2 by MrWho, 11 years ago

Resolution: invalid
Status: newclosed

no log no help

comment:3 by hapeba, 11 years ago

Resolution: invalid
Status: closedreopened

The same for me.
Because everyone can test and see it, a special log-file is needless.

comment:4 by petar258, 11 years ago

about 40 ms higher with 8637 and 8638

by theparasol, 11 years ago

Attachment: tryfix1.patch added

No wait on fd in error, wait max 3000ms for fd to become ready

comment:5 by hapeba, 11 years ago

the patch don't work ...

comment:6 by hapeba, 11 years ago

apply patch tryfix1.patch
patching file oscam.c
Hunk #1 succeeded at 902 (offset 71 lines).
Hunk #2 succeeded at 911 (offset 71 lines).
Hunk #3 succeeded at 995 (offset 71 lines).
patching file oscam-work.c
Hunk #1 succeeded at 106 (offset -2 lines).
Hunk #2 FAILED at 151.
1 out of 2 hunks FAILED -- saving rejects to file oscam-work.c.rej

comment:7 by theparasol, 11 years ago

Forget about tryfix1, it makes it even worse. The whole poll approach seems bad so
revert part of the 8637 commit, hope ecm times returned to normal now.

Last edited 11 years ago by theparasol (previous) (diff)

comment:8 by Admin, 11 years ago

Resolution: fixed
Status: reopenedclosed

Was reverted in r8639 which should have fixed the issues.

Note: See TracTickets for help on using tickets.