Thomas Nau schrieb: > Hi > >> take a look at the following output of iostat -xzn 5: >> extended device statistics >> r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device >> 119.4 0.0 11014.2 0.0 0.0 1.7 0.0 13.9 0 96 md/d10 >> 59.6 0.0 5529.4 0.0 0.0 0.9 0.0 14.4 0 72 md/d11 >> 59.8 0.0 5484.8 0.0 0.0 0.8 0.0 13.4 0 70 md/d12 >> 0.4 169.2 3.2 11143.8 0.0 1.5 0.0 8.6 0 52 md/d20 >> 0.2 169.2 1.6 11143.8 0.0 1.1 0.0 6.5 0 38 md/d21 >> 0.2 169.2 1.6 11143.8 0.0 1.0 0.0 6.2 0 38 md/d22 >> 59.8 173.2 5531.0 11145.8 0.0 2.0 0.0 8.4 0 77 c0t0d0 >> 60.0 169.2 5486.4 11143.8 0.0 1.8 0.0 8.0 0 78 c13t0d0 >> >> d10 consists of d11 and d12 >> d20 consists of d21 and d22 >> >> So how is it possible that reads of d11,d12 sum up in d10, but writes >> of d20, d21, and d22 are all equal. Is this a bug? >> >> This occurred during 'lumake -n slice1' on: >> $ uname -a >> SunOS azalin 5.10 Generic_127111-01 sun4u sparc SUNW,Sun-Blade-2500 >> >> Any comments? > > > Assuming that d10 and d20 are mirror devices with two sub mirrors each... > Reads to such devices are handled round robin by default so the total > read from the metadevice will be the sum of all submirrors. Of course > for writes whatever you dump into d20 shows up on each of the submirrors > > Thomas > > ----------------------------------------------------------------- > GPG fingerprint: B1 EE D2 39 2C 82 26 DA A5 4D E0 50 35 75 9E ED grmph - of course that's it. I guess the week was a little bit too long, or was it the beer that I just enjoyed?
Cheers, Thomas _______________________________________________ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
