1. Good to hear about the patch. 2. If RedHat/gcc/malloc were truly broken, you would think we would have heard the loud screams of 100s of users...
I'd be a lot more suspicious that there is something in ntop, some assumption being made, that's just not right. Still, I've had it blow up on me and looked at the memory w/ gdb - as has the user below, and it looks right. Perhaps gcc/malloc has some problem reusing memory... or something - if I see it again, I'll file up the old pmmm patch and see if it's a free of unallocated memory that somehow it's able to store into... Let's see how the patch goes... -----Burton -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Luca Deri Sent: Tuesday, June 04, 2002 7:59 AM To: [EMAIL PROTECTED] Subject: Re: [Ntop-dev] ntop crash Hi Burton, I've added a patch to ntop for the problem below. I'm currently testing the patch so I will let you know later this week. My feeling is that the Linux memory management (expecially in the latest releases) is somehow broken. I have noticed for instance that in the latest RH 7.3 ntop does not always complete the shutdown (a call to free() loops forever). This didn't happen on 7.2 AFAIK. ntop on Solaris with the very same traffic runs beautifully. Any clue? Cheers, Luca ----- Original Message ----- From: "Burton M. Strauss III" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Monday, June 03, 2002 4:28 PM Subject: RE: [Ntop-dev] ntop crash > Assuming this is 2.0.99RC1, this is the bug we're chasing... don't know > what else I can say - it doesn't seem to fail for Luca or myself. > > Luca did add a semaphore this weekend, so you might try the current cvs > version (call it 2.0.99RC1+) and let us know! > > -----Burton > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf > Of loikatsat > Sent: Friday, May 31, 2002 1:20 AM > To: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Subject: [Ntop-dev] ntop crash > > > > hi, > i need your help because of ntop crash after mins. i use -K to get the > error > message from syslogs: > > ntop caught signal 11 > *****ntop error: Signal(11) > backtrace is: > 1. /lib/i686/libpthread.so.0 [0x402608d5] > 2. /lib/i686/libc.so.6 [0x4053e848] > 3. /lib/i686/libc.so.6(__libc_free+0xb4) [0x40590bf4] > 4. /usr/local/lib/libntop-2.0.so.99(ntop_safefree+0x51) > [0x402b320d] > 5. /usr/local/lib/libntop-2.0.so.99(freeSession+0x25b) > [0x402c029b] > 6. > /usr/local/lib/libntop-2.0.so.99(scanTimedoutTCPSessions+0x1ac) > [0x402c0450] > 7. /usr/local/lib/libntop-2.0.so.99(purgeIdleHosts+0x316) > [0x402af636] > 8. /usr/local/lib/libntop-2.0.so.99(scanIdleLoop+0x65) > [0x402b65c9] > 9. /lib/i686/libpthread.so.0 [0x4025db9c] > 10. /lib/i686/libc.so.6(__clone+0x3a) [0x405fd1ea] > Cleaning up... > Waiting until threads terminate... > Freeing hash host instances... (2 device(s) to save) > > _______________________________________________ > Ntop-dev mailing list > [EMAIL PROTECTED] > http://listgateway.unipi.it/mailman/listinfo/ntop-dev > > _______________________________________________ > Ntop-dev mailing list > [EMAIL PROTECTED] > http://listgateway.unipi.it/mailman/listinfo/ntop-dev > _______________________________________________ Ntop-dev mailing list [EMAIL PROTECTED] http://listgateway.unipi.it/mailman/listinfo/ntop-dev _______________________________________________ Ntop-dev mailing list [EMAIL PROTECTED] http://listgateway.unipi.it/mailman/listinfo/ntop-dev