Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Yes, we have developed a testing couple of servers with vtund, compressed and unencypted tunnel. We have try zlib and lzo compression with level 9 and we got very high speeds even through an Internet link which max throughput was 250KB/s. Only problem we have had is that at such high speeds, on first sync, we have a kernel panic at primary node (it never happened with rate limited at 1M on drbd setup).We think this may be due to kernel compilation stuff as it have happened only when primary node was a 2.6.15 and never when primary was 2.6.8... We decided to wait until 0.8 was released for production use to restart the testing. --- Todd Denniston <Todd.Denniston at ssa.crane.navy.mil> wrote: > Lars, > Would it make sense, in 0.8.x or 0.9.x, that when > doing a full sync, drbd do a > minimal look at the data it is sending and handle > "sparse" (large sequences of > 0's) data 'efficiently' like tar and cp do, i.e., > when drbd notices it is sending 100 4k blocks of > data, instead of sending 400k > of 0, send a "drbd_sparse 100" packet to the peer > which then turns that into > 0's on the media? I think that at this point that would be complicated enough to simply add compression instead. lzo would probably be a great fit for this: http://www.oberhumer.com/opensource/lzo/ but I suspect that this is beyond what the developers want. They would probably tell you to use openvpn. Has anyone ever tried openvpn with drbd without encryption enabled to see if it can speed things up (it can use lzo)? -Martin __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ drbd-user mailing list drbd-user at lists.linbit.com http://lists.linbit.com/mailman/listinfo/drbd-user