Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Looks like it worked. :-) Now I have up2date everything (kernel,
drbd). :-) Just 45 minutes of unscheduled downtime. :-)
And yes, my reiserfs has overwritten metadata (as suspected), while
on the first node with resizing virtual volume fist, it upgraded
metadata no problem. :-)
levinja etc # drbdadm create-md all
v08 Magic number not found
md_offset 161413591040
al_offset 161413558272
bm_offset 161408630784
Found reiser filesystem which uses 157155328 kB
current configuration leaves usable 157625616 kB
==> This might destroy existing data! <==
Do you want to proceed?
[need to type 'yes' to confirm] yes
v07 Magic number not found
v08 Magic number not found
You want me to create a v08 style flexible-size internal meta data
block.
There apears to be a v07 fixed-size internal meta data block
already in place on /dev/vg/sqlmail at byte offset 161279377408
Convert the existing v07 meta-data to v08?
[need to type 'yes' to confirm] yes
Converting meta data...
Writing meta data...
New drbd meta data block sucessfully created.
--== Creating metadata ==--
success
Ignoring additional arguments
v08 Magic number not found
md_offset 322474864640
al_offset 322474831872
bm_offset 322464989184
Found reiser filesystem which uses 314441728 kB
current configuration leaves usable 314907216 kB
==> This might destroy existing data! <==
Do you want to proceed?
[need to type 'yes' to confirm] yes
v07 Magic number not found
v08 Magic number not found
You want me to create a v08 style flexible-size internal meta data
block.
There apears to be a v07 fixed-size internal meta data block
already in place on /dev/vg/web at byte offset 322340651008
Convert the existing v07 meta-data to v08?
[need to type 'yes' to confirm] yes
Converting meta data...
Writing meta data...
New drbd meta data block sucessfully created.
--== Creating metadata ==--
success
levinja ~ # cat /proc/drbd
version: 8.0.11 (api:86/proto:86)
GIT-hash: b3fe2bdfd3b9f7c2f923186883eb9e2a0d3a5b1b build by
root at levinja, 2008-02-20 15:34:53
0: cs:SyncSource st:Primary/Secondary ds:UpToDate/Inconsistent C r---
ns:287452 nr:0 dw:2472 dr:303596 al:45 bm:17 lo:0 pe:14 ua:165 ap:0
[>....................] sync'ed: 0.2% (153650/153931)M
finish: 7:17:03 speed: 5,988 (5,856) K/sec
resync: used:1/31 hits:18099 misses:20 starving:0 dirty:0
changed:20
act_log: used:0/257 hits:573 misses:47 starving:0 dirty:2
changed:45
1: cs:SyncSource st:Primary/Secondary ds:UpToDate/Inconsistent C r---
ns:4535284 nr:0 dw:1380 dr:4543568 al:34 bm:277 lo:18 pe:14 ua:
175 ap:0
[>....................] sync'ed: 0.1% (303110/303382)M
finish: 13:47:41 speed: 6,212 (5,804) K/sec
resync: used:1/31 hits:283359 misses:280 starving:0 dirty:0
changed:280
act_log: used:0/257 hits:311 misses:34 starving:0 dirty:0
changed:34
lev ~ # cat /proc/drbd
version: 8.0.11 (api:86/proto:86)
GIT-hash: b3fe2bdfd3b9f7c2f923186883eb9e2a0d3a5b1b build by root at lev,
2008-02-20 15:57:25
0: cs:SyncTarget st:Secondary/Primary ds:Inconsistent/UpToDate C r---
ns:0 nr:226228 dw:226228 dr:0 al:0 bm:13 lo:1 pe:180 ua:0 ap:0
[>....................] sync'ed: 0.2% (153710/153931)M
finish: 6:53:07 speed: 6,316 (5,952) K/sec
resync: used:2/31 hits:14303 misses:17 starving:0 dirty:0
changed:17
act_log: used:0/257 hits:0 misses:0 starving:0 dirty:0
changed:0
1: cs:SyncTarget st:Secondary/Primary ds:Inconsistent/UpToDate C r---
ns:0 nr:217884 dw:217756 dr:0 al:0 bm:15 lo:5 pe:171 ua:4 ap:0
[>....................] sync'ed: 0.1% (303170/303382)M
finish: 14:52:05 speed: 5,776 (5,728) K/sec
resync: used:1/31 hits:13777 misses:16 starving:0 dirty:0
changed:16
act_log: used:0/257 hits:0 misses:0 starving:0 dirty:0
changed:0
:-)
After it synces i will start heartbeat also on syncTarget.
Sorry for spamming the list. :-)
Regards,
M.
On 2008.2.20, at 14:00, Martin Gombac wrote:
> Addition.
> I might know why id didn't detect metadata. Probably cause my drbd
> device extends thru whole virtual volume.
> Now i'm gonna upgrade the other node too. Turn drbd off, extend
> virtual volume, and recreate metadata. Then ran drbd again.
> On the second node i will also exted virtual volume for 300MB, then
> create-md, then drbadam -- --discard-my-data all connect or somth. ...
> I hope it works. :-)
>
> Regards,
> M.
>
> On 2008.2.20, at 13:27, Martin Gombac wrote:
>
>> Hi,
>>
>> i wanted to upgrade my kernel due to recent security issues, but
>> after kernel upgrade found out that 0.7.24 and 0.7.25 won't
>> compile on 2.6.23 and 2.6.24. I decided i won't manually install
>> it using patches from Igor Zhbanov, but will rather upgrade to
>> 0.8.11, which proved to be a mistake.
>> I have updated my drbd.conf with new options and kept on `on node
>> {}` parts from it intact. Then i wanted to upgrade metadata and
>> continue with second node. But it says it can't find 0.7 version
>> of metadata.
>>
>> What should i do? I successfully upgraded from 0.7.24 (i think) to
>> 0.8.8 on another cluster but have problems with this one.
>>
>> --
>> lev etc # drbdadm create-md web
>>
>> v08 Magic number not found
>> v07 Magic number not found
>> About to create a new drbd meta data block
>> on /dev/vg/web.
>>
>> ==> This might destroy existing data! <==
>>
>> Do you want to proceed?
>> [need to type 'yes' to confirm]
>> lev etc #
>> --
>>
>> For now i'll leave cluster running with just one node. :-(
>>
>> Regards,
>> M.
>