Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
--On Donnerstag, 27. Oktober 2005 09:25 -0500 Tony McCallie <TonyMcCallie at tfchurch.org> wrote: > [root at creativetest2 ~]# cat /proc/drbd > version: 0.7.11 (api:77/proto:74) > SVN Revision: 1807 build by bachbuilder at g39.physik.fu-berlin.de, > 2005-09-30 12:11:16 > 0: cs:SyncTarget st:Secondary/Primary ld:Inconsistent > ns:0 nr:19091224 dw:19091224 dr:0 al:0 bm:12847 lo:0 pe:2556 ua:0 > ap:0 > [=>..................] sync'ed: 10.0% (168266/186909)M > finish: 179:29:02 speed: 252 (288) K/sec > > The only thing I can see in the syslogs that looks like a problem is > from heartbeat: > > (repeated many times) > heartbeat[27637]: 2005/10/27_09:24:32 WARN: 1 lost packet(s) for > [creativetest1.int-tfchurch.org] [93:95] > heartbeat[27637]: 2005/10/27_09:24:32 info: No pkts missing from > creativetest1.int-tfchurch.org! I had similar problems with drbd. The problem was, that the on-board SATA controller and the NIC used the same interrupt, and for some reason the SATA controller generated interrupts at about 15000/s. As soon as we disconnected the harddrive everything worked fine. You should have a look at the number of interrupts generated on the IRQ line used by the drbd NIC (cat /proc/interrupts), perhaps it is something similar. Regards Norbert -- Norbert Langermann norbert at langermann.net OpenPGP-Key 0x51431418 Verschränkung ist, wenn man das eine Teilchen kitzelt und das andere lacht.