Messages in this thread | | | Date | Thu, 21 Jan 1999 23:40:29 +0100 | From | Alessandro Suardi <> | Subject | Re: That whole Netscape problem... |
| |
Neil Conway wrote: > > Linux account wrote: > > > > On Tue, 19 Jan 1999, Charles Shannon Hendrix wrote: > > > The other day I turned Java completely off in 4.08 and it has not > > > crashed for nearly 3 days. This is with kernel 2.1.132. > > > > Indeed. On 2.0.36 and 2.1.anything, I've seen Netscape eat the processor > > and freeze with lots of those SIGALRM calls scrolling away to infinity. > > Turn off Java and, though it hogs and hogs memory, it doesn't get into > > this loop. > > > > So, may I suspect the JVM? I hear a new JVM is coming in 5. > > Sorry to spoil the party but I never use Java (am I paranoid or isn't it > about as secure as a paper bag) and I have regular hangs and crashes > (like two-three per week) on 4.06 under 2.0.3x (2.0.35 at present). > > I do usually have Javascript turned on, but that's about it. I use it > for web and mail. > > The symptoms are either of "Bus Error" (mostly) to just burning CPU and > doing nothing (occasionally). Maybe I should strace() it next time it > gets stuck in a loop, but I haven't yet. >
The SIGALRM loop happens here as well, only thing I noticed it _seems_ to happen when one of the windows attempts a TCP connection to a site which [don't kill me, I have no knowledge of a TCP connection handshaking phase] isn't quite fast in responding and netstat -tn shows the state as SYN_SENT.
Other data comes in for other windows, Recv-Q's build up, but if that TCP connection stays in SYN_SENT there is nothing to do except kill -SEGV the Netscape PID. Kill -TERM can at times make all windows go away but the process stays there, so a further SEGV is needed to complete the job :(
--alessandro <asuardi@uninetcom.it> <asuardi@it.oracle.com>
Linux 2.0.36/2.2.0-pre7 glibc-2.0.7-29 gcc-2.8.1 binutils-2.9.1.0.19a
"I hate bugs which disappear just as soon as you start trying to narrow things down." -- Stephen Tweedie
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |