On Thu, Jul 18, 2013 at 09:26:31AM +0200, Valerio Pachera wrote: > root at test004:~# du -sh /mnt/sheep/meta/cache/ > 18G /mnt/sheep/meta/cache/ > > I shut down the guest, then > > # collie vdi cache flush backup > # collie vdi cache flush backup_data > # collie vdi cache delete backup_data > # collie vdi cache delete backup > > # du -sh /mnt/sheep/meta/cache/ > 17G /mnt/sheep/meta/cache/ > > Notice that, before doing the test of the previous mail, I shutdown > the guest and rerun the guest. > Is it normal that /mnt/sheep/meta/cache/ is still full of data? No. I think the remained cache is something related to copy-on-write mechanism regards of snapshot. Data will be shared between snapshots. If you are sure 1. no guest running 2. sheep still alive and runnign fine You can simply rm -f /mnt/sheep/meta/cache/* to purge the cache. Anyway, collie vdi cache delete support '-s tag' option to delete the desired cache. e.g, yliu at ubuntu-precise:~/sheepdog$ ls /tmp/cache0/cache/ a34a05 a34a06 a34a07 a34a08 yliu at ubuntu-precise:~/sheepdog$ collie/collie vdi list Name Id Size Used Shared Creation time VDI id Copies Tag test 0 20 MB 20 MB 0.0 MB 2013-07-18 14:23 7c2b25 2 s data 1 10 GB 296 MB 0.0 MB 2013-07-18 14:23 a34a05 2 t1 <-- I'll delete a34a05 s data 2 10 GB 164 MB 280 MB 2013-07-18 14:25 a34a06 2 t2 s data 3 10 GB 1008 MB 436 MB 2013-07-18 14:25 a34a07 2 t3 s data 4 10 GB 1.0 GB 1.4 GB 2013-07-18 14:31 a34a08 2 t4 data 0 10 GB 8.0 MB 2.4 GB 2013-07-18 14:39 a34a09 2 yliu at ubuntu-precise:~/sheepdog$ collie/collie vdi cache delete -s t1 data yliu at ubuntu-precise:~/sheepdog$ ls /tmp/cache0/cache/ a34a06 a34a07 a34a08 <--- a34a05 is deleted Thanks Yuan |