Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug
report.
http://bugzilla.quagga.net/show_bug.cgi?id=402
Summary: second zebra process wipes out static routes of the
first one
Product: Quagga
Version: CVS
Platform: Other
OS/Version: other
Status: NEW
Severity: normal
Priority: Medium
Component: zebra
AssignedTo: web@pilot.org.ua
ReportedBy: web@pilot.org.ua
CC: paul@dishone.st,wawa@yandex-team.ru
This bug is currently seen on FreeBSD hosts (5.5 and 6.2) right after OS
startup. If a host is configured to have static routes in zebra.conf and quagga
service enabled, the first zebra process will start and load all the routes
correctly. This could be seen in the logfile with appropriate debug options
configured.
5 or 10 seconds later another zebra process gets started up (the reason isn't
yet found why). Quite soon we get a "Could not lock pid_file" message and the
new process terminates, but right before that the old zebra process gets a
sequence of RTM_DELETE routing messages from the kernel originated by a
different PID with PROTO1 flag set (the new zebra process most probably).
As a result all the static routes configured in zebra.conf are missing from the
kernel and are not marked as '*' in RIB accordingly. The original zebra process
doesn't try to reinstall the routes (should it?). quagga service restart solves
the inconsistency, so the second copy probably originates somewhere from FreeBSD
init scripts.
Of course, having keep_kernel_mode == 0 by default is right, but a more correct
behaviour would be to avoid modifying FIB before we assure that the current
zebra process is the only one (via pidfile locking).
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
_______________________________________________
Quagga-bugs mailing list
Quagga-bugs@lists.quagga.net
http://lists.quagga.net/mailman/listinfo/quagga-bugs
comments should be made in the comments box of this bug
report.
http://bugzilla.quagga.net/show_bug.cgi?id=402
Summary: second zebra process wipes out static routes of the
first one
Product: Quagga
Version: CVS
Platform: Other
OS/Version: other
Status: NEW
Severity: normal
Priority: Medium
Component: zebra
AssignedTo: web@pilot.org.ua
ReportedBy: web@pilot.org.ua
CC: paul@dishone.st,wawa@yandex-team.ru
This bug is currently seen on FreeBSD hosts (5.5 and 6.2) right after OS
startup. If a host is configured to have static routes in zebra.conf and quagga
service enabled, the first zebra process will start and load all the routes
correctly. This could be seen in the logfile with appropriate debug options
configured.
5 or 10 seconds later another zebra process gets started up (the reason isn't
yet found why). Quite soon we get a "Could not lock pid_file" message and the
new process terminates, but right before that the old zebra process gets a
sequence of RTM_DELETE routing messages from the kernel originated by a
different PID with PROTO1 flag set (the new zebra process most probably).
As a result all the static routes configured in zebra.conf are missing from the
kernel and are not marked as '*' in RIB accordingly. The original zebra process
doesn't try to reinstall the routes (should it?). quagga service restart solves
the inconsistency, so the second copy probably originates somewhere from FreeBSD
init scripts.
Of course, having keep_kernel_mode == 0 by default is right, but a more correct
behaviour would be to avoid modifying FIB before we assure that the current
zebra process is the only one (via pidfile locking).
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
_______________________________________________
Quagga-bugs mailing list
Quagga-bugs@lists.quagga.net
http://lists.quagga.net/mailman/listinfo/quagga-bugs