sks-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Sks-devel] Stack overflow?


From: Yaron M. Minsky
Subject: Re: [Sks-devel] Stack overflow?
Date: Tue, 02 Dec 2003 15:46:48 -0500

native-code for production.  The only advantage of bytecode in this case
is that you can get stack trace information.  But that's only available
when an exception actually kills the application, and SKS catches most
exceptions, so no stack traces are available anyway.

y

On Tue, 2003-12-02 at 15:23, Sacha J. Bernstein wrote:
> I'm running the bytecode version from the latest CVS.  I was running 
> debuglevel of 3.
> 
> I just restarted with a debuglevel of 5, and the recon log is below. 
> Nothing odd is in the db log at all, just a bunch of hashqueries from my 
> gossip peers.
> 
> Am I better off running the bytecode versions or not for a production 
> server?  Should it be bytecode for debug and native-code for production? 
> Again, I'm not at all familiar with ocaml...
> 
> Thanks,
> 
> Sacha
> 
> 
> 
> 2003-12-02 15:07:13 Opening log
> 2003-12-02 15:07:13 sks_recon, SKS version 1.0.6
> 2003-12-02 15:07:13 Copyright Yaron Minsky 2002-2003
> 2003-12-02 15:07:13 Licensed under GPL.  See COPYING file for details
> 2003-12-02 15:07:13 recon port: 11370
> 2003-12-02 15:07:13 Opening PTree database
> 2003-12-02 15:07:13 Setting up PTree data structure
> 2003-12-02 15:07:13 PTree setup complete
> 2003-12-02 15:07:13 Initiating catchup
> 2003-12-02 15:07:13 Fetching filters
> 2003-12-02 15:07:13 Starting event loop
> 2003-12-02 15:08:15 Membership: <ADDR_INET 
> 134.169.171.249:11370>(sks.keyserver.penguin.de 11370       ), <ADDR_INET 
> 134.169.171.253:11370>(calvin.lk.etc.tu-bs.de 11370), <ADDR_INET 
> 62.245.184.24:11370>(keyserver.noreply.org 11370), <ADDR_INET 
> 68.91.150.38:11370>(dannyj.dynip.com 11370), <ADDR_INET 
> 213.141.74.169:11370>(keys.se.linux.org 11370)
> 2003-12-02 15:08:15 Recon partner: <ADDR_INET 134.169.171.249:11370>
> 2003-12-02 15:08:15 Initiating reconciliation
> 2003-12-02 15:08:20 <recon as client> error in callback.: 
> Sys_error("Connection reset by peer")
> 2003-12-02 15:08:45 Beginning recon as server, client: <ADDR_INET 
> 213.141.74.169:42503>
> 2003-12-02 15:08:45 Joining reconciliation
> 2003-12-02 15:14:13 Reconciliation complete
> 2003-12-02 15:14:16 <reconciliation handler> error in callback.: Stack 
> overflow
> 2003-12-02 15:14:16 Beginning recon as server, client: <ADDR_INET 
> 68.91.150.38:4179>
> 2003-12-02 15:14:16 Joining reconciliation
> 2003-12-02 15:19:09 Reconciliation complete
> 2003-12-02 15:19:12 <reconciliation handler> error in callback.: Stack 
> overflow
> 
> 
> 
> 
> 
> 
> --On Tuesday, December 02, 2003 14:53:44 -0500 "Yaron M. Minsky" 
> <address@hidden> wrote:
> 
> > Don't apologize.  As long as we actually find and remove a bug, I'm
> > happy.
> >
> > A few questions: are you running the byte-code (i.e., the ones that end
> > in .bc) or native-code versions?  Which version are you running?  Is it
> > the latest CVS or is it the 1.0.5 release?  What debugging level are you
> > running?  Try setting debuglevel to 5 in the log file and see if you get
> > the error again.
> >
> > Yaron
> >
> > On Tue, 2003-12-02 at 11:43, Sacha J. Bernstein wrote:
> >> I hate to be the new guy causing all the trouble, but I'm getting the
> >> following lines in my log.recon occationally.  I'm alright at debugging
> >> C,  but I've never used Caml before, and I have no idea what I'm doing
> >> with it.  What can I do that will give you enough information to find
> >> the problem (if  there is one)?
> >>
> >> Also, a feature request (although there may be reasons to not do this)
> >> is  to log the IP address of the clients connecting to the sks server.
> >> I  suppose this is bad from a privacy standpoint, but maybe there could
> >> be a  way to enable it for debugging only?
> >>
> >> 2003-12-02 11:08:13 <reconciliation handler> error in callback.: Stack
> >> overflow
> >> 2003-12-02 11:08:13 <reconciliation handler> error in callback.:
> >> End_of_file 2003-12-02 11:08:59 <recon as client> callback timed out.
> >> 2003-12-02 11:09:00 <reconciliation handler> error in callback.:
> >> End_of_file 2003-12-02 11:14:48 <reconciliation handler> error in
> >> callback.: Stack  overflow
> >> 2003-12-02 11:14:50 <reconciliation handler> error in callback.:
> >> End_of_file 2003-12-02 11:22:47 <recon as client> error in callback.:
> >> Stack overflow 2003-12-02 11:28:33 <reconciliation handler> error in
> >> callback.: Stack  overflow
> >> 2003-12-02 11:29:49 <recon as client> error in callback.:
> >> Sys_error("Connection reset by peer")
> >>
> >> Thanks,
> >>
> >> Sacha
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> Sks-devel mailing list
> >> address@hidden
> >> http://mail.nongnu.org/mailman/listinfo/sks-devel
> > --
> >| --------/            Yaron M. Minsky              \--------|
> >| --------\ http://www.cs.cornell.edu/home/yminsky/ /--------|
> >
> > Open PGP --- KeyID B1FFD916
> > Fingerprint: 5BF6 83E1 0CE3 1043 95D8 F8D5 9F12 B3A9 B1FF D916
> >
> >
> 
-- 
|--------/            Yaron M. Minsky              \--------|
|--------\ http://www.cs.cornell.edu/home/yminsky/ /--------|

Open PGP --- KeyID B1FFD916
Fingerprint: 5BF6 83E1 0CE3 1043 95D8 F8D5 9F12 B3A9 B1FF D916






reply via email to

[Prev in Thread] Current Thread [Next in Thread]