Just realized we can make checkpoint anytime too, so we should be able
to resolve the corrupt database with "crash_dump" without losing any
data, right?
On 17/09/15 10:47, Balazs Varhegyi wrote:
Hi,
We did an integrity check for the first time with " backup
'/dev/null'" and it stopped with:
19:41:49 Dumping the schema tables
19:41:49 Dumping the registry
19:41:49 Dumping the schema done
21:58:12 * Monitor: High disk read (2)
22:00:54 * Monitor: High disk read (2)
....
01:07:34 * Monitor: High disk read (2)
01:08:14 /usr/bin/virtuoso-t() [0x8d4afd]
01:08:14 /usr/bin/virtuoso-t() [0x8d4b78]
...
01:08:14 /usr/bin/virtuoso-t() [0x5db66c]
01:08:14 /usr/bin/virtuoso-t(sf_sql_execute_w+0x74) [0x5db884]
01:08:14 /usr/bin/virtuoso-t() [0x8d8f38]
01:08:14 /usr/bin/virtuoso-t() [0x8df6fc]
01:08:14 /lib/x86_64-linux-gnu/libpthread.so.0(+0x8182) [0x7fa9b54d7182]
01:08:14 /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fa9b4ae147d]
01:08:14 GPF: rltrx.c:1699 mismatched lt thread counts in lt_transact
but virtuoso seems operational, it can be stopped/started, we can
query the data.
Should we worry about it?
Documentation says: " If the server crashes before completing the
backup process, then the database is indeed corrupt and needs to be
recovered"
Doc also suggests to make a crash dump but says "The database will
contain the transactions that were committed as of the last successful
checkpoint.".
This database has been running for 6 months already.
Does it mean if we do the crash dump and the last successful
checkpoint was 5 months ago, then we lose 5 months of data?
Regards,
Balazs Varhegyi
------------------------------------------------------------------------------
_______________________________________________
Virtuoso-users mailing list
Virtuoso-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/virtuoso-users
