[sheepdog-users] Snapshot and Cache stress test

[Valerio Pachera]

I tried the same with cache=directsync on the guest.
I get almost the same result:

Aug 19 16:57:50  ERROR [gway 31433] wait_forward_request(212) fail
4fd26d00000003, No object found
Aug 19 16:57:54  EMERG [main] crash_handler(250) sheep exits
unexpectedly (Segmentation fault).

But sheep daemon is still running!

dog node list hangs up.
qemu is still running (guest shows I/O error) and I have to use kill
-9 to stop it.

More messages in sheep.log

---
...
Aug 19 16:57:26  ERROR [gway 31387] gateway_read_obj(62) local read
804fd25b00000000 failed, No object found
Aug 19 16:57:26  ERROR [gway 31384] gateway_read_obj(62) local read
804fd25d00000000 failed, No object found
Aug 19 16:57:26  ERROR [gway 31384] gateway_read_obj(62) local read
804fd26000000000 failed, No object found
Aug 19 16:57:26  ERROR [gway 31387] gateway_read_obj(62) local read
804fd26100000000 failed, No object found
Aug 19 16:57:26  ERROR [gway 31356] gateway_read_obj(62) local read
804fd26200000000 failed, No object found
Aug 19 16:57:26  ERROR [gway 31384] gateway_read_obj(62) local read
804fd26300000000 failed, No object found
Aug 19 16:57:47  ERROR [net 31414] do_read(220) connection is closed
(48 bytes left)
Aug 19 16:57:47  ERROR [net 31414] rx_work(537) failed to read a header
Aug 19 16:57:47  ERROR [net 31326] do_read(220) connection is closed
(48 bytes left)
Aug 19 16:57:47  ERROR [net 31326] rx_work(537) failed to read a header
Aug 19 16:57:48  ERROR [net 30537] do_read(220) connection is closed
(48 bytes left)
Aug 19 16:57:48  ERROR [net 30537] rx_work(537) failed to read a header
Aug 19 16:57:48  ERROR [net 31414] do_read(220) connection is closed
(48 bytes left)
Aug 19 16:57:48  ERROR [net 31414] rx_work(537) failed to read a header
Aug 19 16:57:48  ERROR [net 31414] do_read(220) connection is closed
(48 bytes left)
Aug 19 16:57:48  ERROR [net 31414] rx_work(537) failed to read a header
...
---