poor performance, error messages, crashes

RESOLVED INCOMPLETE

Status

RESOLVED INCOMPLETE
6 years ago
6 years ago

People

(Reporter: allltaken, Unassigned)

Tracking

SeaMonkey 2.14 Branch
x86
Linux

Firefox Tracking Flags

(Not tracked)

Details

(Reporter)

Description

6 years ago
User Agent: Mozilla/5.0 (X11; Linux i686; rv:16.0) Gecko/20121026 Firefox/16.0 SeaMonkey/2.13.2
Build ID: 20121026204658

Steps to reproduce:

I ran several versions of Seamonkey in a terminal, browsed my usual websites (Yahoo, NY Times, Huffington Post) and looked for performance issues and error messages. The versions were 2.12.1, 2.13.2, 2.14, 2.15a2 and 2.17a1.




Actual results:

I found no errors reported from 2.12.1. I found one error from 2.13.2 that happened while my system was stressed running a daily security check. I found repeated error messages from the later versions during normal runtime conditions. I have also found a jump in memory demands of 2.14 compared to 2.12.1, and an unsatisfactory rate of performance degradation (system hanging possibly during garbage collection), very slow keyboard handling, minimizing, maximizing and some crashing in 2.14 and later versions.

I should add that I'm using a Mandriva Linux 2010.0 system with 512Mb of main memory, 3Gb of swap, and an Athlon XP processor with a label suggesting it's running at 1.9Ghz. It's connected to DSL.

Here are samples of the error messages. All except the sample for SM 2.13.2 were repeated many times.

Terminal error msg from SM213 while system security check was running:
WARNING:  Unable to create pipe named "4664.83fdda80.337240514"  in server mode error (Too many open files).: file /builds/slave/rel-comm-rel-lnx-bld/build/mozilla/ipc/chromium/src/chrome/common/ipc_channel_posix.cc,  line  265
###! ! ! ABORT:  file /builds/slave/rel-comm-rel-lnx-bld/build/mozilla/ipc/chromium/src/base/rand_util_posix.cc, line 19
###! ! ! ABORT:  file /builds/slave/rel-comm-rel-lnx-bld/build/mozilla/ipc/chromium/src/base/rand_util_posix.cc, line 19
ExceptionHandler::GenerateDump										sys_pipe failed:Too many open files

Terminal error msg from SM214:
*** NSPlugin Wrapper *** ERROR: NPP_New() wait for reply: Connection closed
*** NSPlugin Wrapper *** WARNING:(../src/npw-wrapper.c:1858):invoke_NPP_Destroy: assertion failed: (rpc_method_invoke_possible(plugin->connection))


Terminal error msg from SM215:
*** NSPlugin Wrapper *** ERROR: NPP_New() wait for reply: Connection closed
*** NSPlugin Wrapper *** WARNING:(../src/npw-wrapper.c:1858):invoke_NPP_Destroy: assertion failed: (rpc_method_invoke_possible(plugin->connection))
NOTE: child process received `Goodbye', closing down

Terminal error msg from SM217:
*** NSPlugin Wrapper *** ERROR: NPP_New() wait for reply: Connection closed
*** NSPlugin Wrapper *** WARNING:(../src/npw-wrapper.c:1858):invoke_NPP_Destroy: assertion failed: (rpc_method_invoke_possible(plugin->connection))



Expected results:

All versions should have worked perfectly with no error messages :-)
(Reporter)

Updated

6 years ago
Summary: poor performance, error messages → poor performance, error messages, crashes

Comment 1

6 years ago
Looks like problem in core ipc code, do you have any plugins installed? If yes, can you disable them for test purposes.
Also provide corresponding crash ids from about:crashes
(Reporter)

Comment 2

6 years ago
Plugins: Mozplugger 1.14.6, Adobe Reader 9.5, Shockwave Flash 11.2, Iced Tea NPR plugin 1.8.7, Shockwave Flash 9.0, mplayer-plugin 3.55, DjView 4.4, Helix DNA plugin 0.4.0.622

about:crashes didn't list any crashes.

FWIW, the crashes I've had recently didn't save any crash data or send reports.

I'm presently using version 2.9.1 which doesn't seem to have any of the performance or crash problems.

Comment 3

6 years ago
Is you still see this problem?
Flags: needinfo?(jrt)
Whiteboard: closeme INCO 2013-06-01

Comment 4

6 years ago
Resolved per whiteboard
Status: UNCONFIRMED → RESOLVED
Last Resolved: 6 years ago
Flags: needinfo?(jrt)
Resolution: --- → INCOMPLETE
Whiteboard: closeme INCO 2013-06-01
You need to log in before you can comment on or make changes to this bug.