Hi, Our VPN runs okay for some time, and every now and then one of the links (we have three hosts connected redundantly) just falls out with no reason. In the syslog I get several (strange?) messages: Feb 14 09:56:02 ophar tinc.pirnet[2816]: 212.187.121.188 port 655: Connection timed out Feb 14 09:56:02 ophar tinc.pirnet[2816]: Could not set up a meta connection to 212.187.121.188 Feb 14 09:56:02 ophar tinc.pirnet[2816]: Still failed to connect to other, will retry in 900 seconds Feb 14 09:56:02 ophar tinc.pirnet[2816]: Error while reading from ethertap device: No buffer space available Feb 14 09:56:02 ophar tinc.pirnet[2816]: Metadata socket read error for vloch_pir (131.155.229.147): No buffer space available Feb 14 09:56:02 ophar tinc.pirnet[2816]: Sending meta data to vloch_pir (131.155.229.147) failed: Bad file descriptor Feb 14 09:56:32 ophar tinc.pirnet[2816]: Bogus data received from unknown (10.1.2.1) Feb 14 09:56:32 ophar tinc.pirnet[2816]: Metadata socket read error for unknown (10.1.2.1): Bad file descriptor Feb 14 09:56:47 ophar dhcpd: fallback_discard: Connection refused Feb 14 09:58:22 ophar tinc.pirnet[2816]: Bogus data received from unknown (10.1.2.1) Could someone give some commentary on those messages? TIA Bart Friederichs ----------------------------------------------------------------------- The internet is a too slow way of doing things you'd never do without it. Bart Friederichs, 1998 ------------------------------------------------------------------------- - Tinc: Discussion list about the tinc VPN daemon Archive: http://mail.nl.linux.org/lists/ Tinc site: http://ftp.nl.linux.org/pub/linux/tinc/
On Wed, Feb 14, 2001 at 10:02:56AM +0100, Bart Friederichs wrote:> Our VPN runs okay for some time, and every now and then one of the links (we > have three hosts connected redundantly) just falls out with no reason. In the > syslog I get several (strange?) messages:[...]> Could someone give some commentary on those messages?Ouch, that are some serious errors. Could you do: ls -lR /proc/`cat /var/run/tinc.pirnet.pid` when this happens again and tell me what you see? If you see really lots of sockets in that directory listing then perhaps we've got a filedescriptor leak. If not, my next best guess would be that your kernel runs out of memory... Could you tell me which version of tinc and the kernel you are using, and on what hardware you're running this? ------------------------------------------- Met vriendelijke groet / with kind regards, Guus Sliepen <guus@sliepen.warande.net> ------------------------------------------- See also: http://tinc.nl.linux.org/ http://www.kernelbench.org/ ------------------------------------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available Url : http://brouwer.uvt.nl/pipermail/tinc/attachments/20010214/6baec605/attachment.pgp