Balance in progress (over 100%)

[ronlaws]

Is this normal?   I'm starting to worry something may be up.

[ronlaws · Accepted solution]

Just an update on this: 

Having now replaced all 4 drives in the array with new ones, the last re-balance i did only took a couple of hours, and did not go over 100%. 

admittedly it only had to balance about 5 blocks, which took a while but at least it didn't drift off in to another dimension this time. 

Currently doing a scrub which is also taking its time, but with any luck the issue seems to be resolved now. 

[StephenB]

It's not something I've seen before.  How long has the balance been running?

Also what firmware is running on the NAS?

[ronlaws]

It's been running overnight, it's running ReadyNAS OS 6.5.2. A sceduled Scrub also started at midnight, which i cancled this morning since this process was still ongoing.. 

Edit: Just checked to see if the process was still running, it appears to be? 

root@ReadyNAS:~# ps aux | grep "btrfs"
root      1158  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-worker]
root      1182  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-worker-hi]
root      1189  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-delalloc]
root      1194  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-flush_del]
root      1206  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-cache]
root      1210  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-submit]
root      1220  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-fixup]
root      1225  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-endio]
root      1236  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-endio-met]
root      1244  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-endio-met]
root      1249  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-endio-rai]
root      1261  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-endio-rep]
root      1274  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-rmw]
root      1277  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-endio-wri]
root      1286  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-freespace]
root      1295  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-delayed-m]
root      1302  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-readahead]
root      1310  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-qgroup-re]
root      1317  0.0  0.0      0     0 ?        S<   Sep24   0:00 [btrfs-extent-re]
root      1828  0.0  0.0      0     0 ?        S    Sep24   0:00 [btrfs-cleaner]
root      1829  0.1  0.0      0     0 ?        D    Sep24   1:35 [btrfs-transacti]
root     14276  1.8  0.0   3204     8 ?        D<   Sep24  20:12 btrfs balance start -dusage 64 -musage 64 /data
root     32204  0.0  0.3   4216  1924 pts/0    S+   12:23   0:00 grep btrfs

root@ReadyNAS:~# btrfs balance status -v /data
Balance on '/data' is running
102 out of about 33 chunks balanced (159 considered), -209% left
Dumping filters: flags 0x7, state 0x1, force is off
  DATA (flags 0x2): balancing, usage=64
  METADATA (flags 0x2): balancing, usage=64
  SYSTEM (flags 0x2): balancing, usage=64

[mdgm-ntgr]

That's not normal.

Can you provide your logs (see the Sending Logs link in my sig)?

[ronlaws]

Logs sent. Also noticed one of the HDD's has a bad sector count (the odd Toshiba one)

[mdgm-ntgr]

I wonder whether this may be related to one of the disks failing.

Do you have a backup?

[ronlaws]

Yes i have a backup, I'd actually recently moved all the data across from an old NV+ though i am in the process of replacing the drives. Once all the drives are replaced with new ones i'll re-run the balance to see if it does it again. If so im preparing myself to do a factory reset.

[ronlaws · Accepted solution]

Just an update on this: 

Having now replaced all 4 drives in the array with new ones, the last re-balance i did only took a couple of hours, and did not go over 100%. 

admittedly it only had to balance about 5 blocks, which took a while but at least it didn't drift off in to another dimension this time. 

Currently doing a scrub which is also taking its time, but with any luck the issue seems to be resolved now. 

[StephenB]

Great news, thx for sharing it.

Scrubs will take a while, since all the files are checked.