Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
To Many people it seems to escape them that writing into the share while doing a fullsync will make the sync slower for obvious reasons (akin to a raid array scrub procedure). El jue, 11-03-2004 a las 13:20, Mark Hellman escribió: > Two servers both with 1 Gigabit NICs and connected by a crossed cable. > When sync starts, the speed is normal (30000 K/sec). After a few minutes, > it drops to the values seen bellow. > > ================================================================= > # cat /proc/drbd > version: 0.6.11 (api:64/proto:62) > > 0: cs:SyncingAll st:Primary/Secondary ns:104102280 nr:0 dw:30380 > dr:104095280 pe:45 ua:0 > [==============>.....] sync'ed: 74.7% (34538/136187)M > finish: 3:41:09h speed: 2,676 (5,653) K/sec > ================================================================= > > /dev/nb0 is 133GB in size (LVM beneath DRBD). If the 30000 K/sec transfer > rate was sustained, the full sync would take about 1.5 hours. But in this > case it took 10 hours! > > Is this a problem of DRBD or aren't IDE hard drives capable of keeping up > sustained high transfer rates? > > > Thanks > Mark > > > _______________________________________________ > drbd-user mailing list > drbd-user at lists.linbit.com > http://lists.linbit.com/mailman/listinfo/drbd-user >