Greg Lehey wrote in message ID
<[EMAIL PROTECTED]>:
> Again, news for me. You can stop a revive and restart it at a later
> time; the state information includes the revive pointer. But the only
> real answer to this one would be to see the statistics.
Well, I deleted the plex and tried again:
root@ohno:~> cat /root/vinum_conf
plex name vol01.p1 org concat
sd name vol01.p1.s0 length 0 drive d1
root@ohno:~> vinum create /root/vinum_conf
2 drives:
D d0 State: up Device /dev/ad4f Avail: 0/35669 MB (0%)
D d1 State: up Device /dev/ad6f Avail: 0/35669 MB (0%)
1 volumes:
V vol01 State: up Plexes: 1 Size: 34 GB
2 plexes:
P vol01.p0 C State: up Subdisks: 1 Size: 34 GB
P vol01.p1 C State: up Subdisks: 1 Size: 34 GB
2 subdisks:
S vol01.p0.s0 State: up PO: 0 B Size: 34 GB
S vol01.p1.s0 State: up PO: 0 B Size: 34 GB
root@ohno:~> vinum attach vol01.p1 vol01
Can't attach vol01.p1 to vol01: a (1)
thats weird
root@ohno:~> vinum printconfig
# Vinum configuration of ohno.home.in-addr.com, saved at Thu Dec 30 08:04:40 1999
drive d0 device /dev/ad4f
drive d1 device /dev/ad6f
volume vol01
plex name vol01.p0 org concat vol vol01
plex name vol01.p1 org concat vol vol01
sd name vol01.p0.s0 drive d0 plex vol01.p0 len 73051272s driveoffset 265s plexoffset 0s
sd name vol01.p1.s0 drive d1 plex vol01.p1 len 73051272s driveoffset 265s plexoffset 0s
root@ohno:~> vinum l -v
2 drives:
Drive d0: Device /dev/ad4f
Created on ohno.home.in-addr.com at Fri Dec 24 10:08:55 1999
Config last updated Thu Dec 30 08:04:29 1999
Size: 37402386944 bytes (35669 MB)
Used: 37402386944 bytes (35669 MB)
Available: 0 bytes (0 MB)
State: up
Last error: none
Active requests: 0
Maximum active: 10
Drive d1: Device /dev/ad6f
Created on ohno.home.in-addr.com at Sat Dec 25 16:39:24 1999
Config last updated Thu Dec 30 08:04:29 1999
Size: 37402386944 bytes (35669 MB)
Used: 37402386944 bytes (35669 MB)
Available: 0 bytes (0 MB)
State: up
Last error: none
Active requests: 0
Maximum active: 2
1 volumes:
Volume vol01: Size: 37402251264 bytes (35669 MB)
State: up
Flags: open
2 plexes
Read policy: round robin
2 plexes:
Plex vol01.p0: Size: 37402251264 bytes (35669 MB)
Subdisks: 1
State: up
Organization: concat
Part of volume vol01
Plex vol01.p1: Size: 37402251264 bytes (35669 MB)
Subdisks: 1
State: faulty
Organization: concat
Part of volume vol01
2 subdisks:
Subdisk vol01.p0.s0:
Size: 37402251264 bytes (35669 MB)
State: up
Plex vol01.p0 at offset 0 (0 B)
Drive d0 (/dev/ad4f) at offset 135680 (132 kB)
Subdisk vol01.p1.s0:
Size: 37402251264 bytes (35669 MB)
State: stale
Plex vol01.p1 at offset 0 (0 B)
Drive d1 (/dev/ad6f) at offset 135680 (132 kB)
root@ohno:~> vinum resetstats
root@ohno:~> vinum lp -s
vol01.p0 0 0 0 0 0
0 0
vol01.p1 0 0 0 0 0
0 0
root@ohno:~> vinum start vol01.p1
Reviving vol01.p1.s0 in the background
root@ohno:~> vinum[49356]: reviving vol01.p1.s0
Worked that time. Dunno what the difference was.
root@ohno:~> vinum l -v
2 drives:
Drive d0: Device /dev/ad4f
Created on ohno.home.in-addr.com at Fri Dec 24 10:08:55 1999
Config last updated Thu Dec 30 08:09:15 1999
Size: 37402386944 bytes (35669 MB)
Used: 37402386944 bytes (35669 MB)
Available: 0 bytes (0 MB)
State: up
Last error: none
Active requests: 1
Maximum active: 10
Drive d1: Device /dev/ad6f
Created on ohno.home.in-addr.com at Sat Dec 25 16:39:24 1999
Config last updated Thu Dec 30 08:09:15 1999
Size: 37402386944 bytes (35669 MB)
Used: 37402386944 bytes (35669 MB)
Available: 0 bytes (0 MB)
State: up
Last error: none
Active requests: 0
Maximum active: 2
1 volumes:
Volume vol01: Size: 37402251264 bytes (35669 MB)
State: up
Flags:
2 plexes
Read policy: round robin
2 plexes:
Plex vol01.p0: Size: 37402251264 bytes (35669 MB)
Subdisks: 1
State: up
Organization: concat
Part of volume vol01
Plex vol01.p1: Size: 37402251264 bytes (35669 MB)
Subdisks: 1
State: faulty
Organization: concat
Part of volume vol01
2 subdisks:
Subdisk vol01.p0.s0:
Size: 37402251264 bytes (35669 MB)
State: up
Plex vol01.p0 at offset 0 (0 B)
Drive d0 (/dev/ad4f) at offset 135680 (132 kB)
Subdisk vol01.p1.s0:
Size: 37402251264 bytes (35669 MB)
State: reviving
Plex vol01.p1 at offset 0 (0 B)
Revive pointer: 7344 kB (0%)
Revive blocksize: 16 kB
Revive interval: 0 seconds
Drive d1 (/dev/ad6f) at offset 135680 (132 kB)
root@ohno:~> vinum ls -sv
Subdisk vol01.p0.s0:
Size: 37402251264 bytes (35669 MB)
State: up
Plex vol01.p0 at offset 0 (0 B)
Drive d0 (/dev/ad4f) at offset 135680 (132 kB)
Reads: 5896
Bytes read: 96600064 (92 MB)
Average read: 16384 bytes
Writes: 0
Bytes written: 0 (0 B)
Subdisk vol01.p1.s0:
Size: 37402251264 bytes (35669 MB)
State: reviving
Plex vol01.p1 at offset 0 (0 B)
Revive pointer: 184 MB (0%)
Revive blocksize: 16 kB
Revive interval: 0 seconds
Drive d1 (/dev/ad6f) at offset 135680 (132 kB)
Reads: 5895
Bytes read: 96583680 (92 MB)
Average read: 16384 bytes
Writes: 11791
Bytes written: 193183744 (184 MB)
Average write: 16384 bytes
As you can see, the revive pointer is at 184MB, but only 92MB has been
read from the `good' plex, the other 92MB has been taken from the
`bad' plex. It really looks like its round robining the reads like it
would if it was a fully silvered mirror.
The full config and stats:
root@ohno:~> vinum printconfig
# Vinum configuration of ohno.home.in-addr.com, saved at Thu Dec 30 08:22:19 1999
drive d0 device /dev/ad4f
drive d1 device /dev/ad6f
volume vol01
plex name vol01.p0 org concat vol vol01
plex name vol01.p1 org concat vol vol01
sd name vol01.p0.s0 drive d0 plex vol01.p0 len 73051272s driveoffset 265s plexoffset 0s
sd name vol01.p1.s0 drive d1 plex vol01.p1 len 73051272s driveoffset 265s plexoffset 0s
root@ohno:~> vinum l -sv
2 drives:
Drive d0: Device /dev/ad4f
Created on ohno.home.in-addr.com at Fri Dec 24 10:08:55 1999
Config last updated Thu Dec 30 08:09:15 1999
Size: 37402386944 bytes (35669 MB)
Used: 37402386944 bytes (35669 MB)
Available: 0 bytes (0 MB)
State: up
Last error: none
Active requests: 0
Maximum active: 10
Reads: 17690
Bytes read: 289832960 (276 MB)
Average read: 16384 bytes
Writes: 0
Bytes written: 0 (0 B)
Drive d1: Device /dev/ad6f
Created on ohno.home.in-addr.com at Sat Dec 25 16:39:24 1999
Config last updated Thu Dec 30 08:09:15 1999
Size: 37402386944 bytes (35669 MB)
Used: 37402386944 bytes (35669 MB)
Available: 0 bytes (0 MB)
State: up
Last error: none
Active requests: 1
Maximum active: 2
Reads: 17689
Bytes read: 289816576 (276 MB)
Average read: 16384 bytes
Writes: 35379
Bytes written: 579649536 (552 MB)
Average write: 16384 bytes
1 volumes:
Volume vol01: Size: 37402251264 bytes (35669 MB)
State: up
Flags:
2 plexes
Read policy: round robin
Reads: 35380
Recovered: 0
Bytes read: 579665920 (552 MB)
Average read: 16384 bytes
Writes: 0
Bytes written: 0 (0 B)
Active requests: 1
2 plexes:
Plex vol01.p0: Size: 37402251264 bytes (35669 MB)
Subdisks: 1
State: up
Organization: concat
Part of volume vol01
Reads: 17690
Bytes read: 289832960 (276 MB)
Average read: 16384 bytes
Writes: 0
Bytes written: 0 (0 B)
Plex vol01.p1: Size: 37402251264 bytes (35669 MB)
Subdisks: 1
State: faulty
Organization: concat
Part of volume vol01
Reads: 17689
Bytes read: 289816576 (276 MB)
Average read: 16384 bytes
Writes: 0
Bytes written: 0 (0 B)
2 subdisks:
Subdisk vol01.p0.s0:
Size: 37402251264 bytes (35669 MB)
State: up
Plex vol01.p0 at offset 0 (0 B)
Drive d0 (/dev/ad4f) at offset 135680 (132 kB)
Reads: 17690
Bytes read: 289832960 (276 MB)
Average read: 16384 bytes
Writes: 0
Bytes written: 0 (0 B)
Subdisk vol01.p1.s0:
Size: 37402251264 bytes (35669 MB)
State: reviving
Plex vol01.p1 at offset 0 (0 B)
Revive pointer: 552 MB (1%)
Revive blocksize: 16 kB
Revive interval: 0 seconds
Drive d1 (/dev/ad6f) at offset 135680 (132 kB)
Reads: 17689
Bytes read: 289816576 (276 MB)
Average read: 16384 bytes
Writes: 35379
Bytes written: 579649536 (552 MB)
Average write: 16384 bytes
Any ideas?
Thanks,
Gary
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
