Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
2010/12/29 Marco Barbero <marco.barbero at gmail.com>: > Hi all. > I'm exeriencing a nasty kernel soft lockup on one cluster. Have to > say I have tons of clusters using same config and all is working fine > > To the point: > > kernel 2.6.36.2 > drbd 8.3.8.1 > (it happens on 2.6.32 and drbd 8.3.7 too) > > On the primary node it happens call tracing task blah blah blocked for > more than xx seconds... etc. > After investigating I have noticed that if I stop drbd on the > secondary node, then the hanging node immediately works fine > again.... > Maybe it's related to nic used that is a RTL-8169...? some updates. I Have replaced rtl with e1000 but no luck in short: - start drbd on node A, become master - mount device, start a service (for example mysql with db on the drbd mounted partition or squid with logs on mounted partition) - start drbd on node B, it becomes secondary stop mysql, all is ok start mysql, all is ok reboot node b (secondary) when it comes up again, I start drbd on it and it becomes again secondary at this point...: stop mysql on node A (primary) and... soft lockup. it hangs until I stop drbd on the node b (secondary). as soon as I stop it, node A works again and mysql (or any other service hanging) stop immediately. I'm stuck now... Any hints please? Thanks in advance I'm using protocol C, one resource, external metadata. Partition is 360GB First time I have a similar issue.. >