summaryrefslogtreecommitdiffstats
path: root/cmd/ztest
diff options
context:
space:
mode:
authorMatthew Ahrens <[email protected]>2019-06-10 11:48:42 -0700
committerBrian Behlendorf <[email protected]>2019-06-10 11:48:42 -0700
commitb8738257c2607c73c731ce8e0fd73282b266d6ef (patch)
treed73e90809b9f413b8894d8ee7fc6ad8a11bff7fe /cmd/ztest
parent5a902f5aaa1fbf6f7e459ec29f6d1d988ec78b0a (diff)
make zil max block size tunable
We've observed that on some highly fragmented pools, most metaslab allocations are small (~2-8KB), but there are some large, 128K allocations. The large allocations are for ZIL blocks. If there is a lot of fragmentation, the large allocations can be hard to satisfy. The most common impact of this is that we need to check (and thus load) lots of metaslabs from the ZIL allocation code path, causing sync writes to wait for metaslabs to load, which can take a second or more. In the worst case, we may not be able to satisfy the allocation, in which case the ZIL will resort to txg_wait_synced() to ensure the change is on disk. To provide a workaround for this, this change adds a tunable that can reduce the size of ZIL blocks. External-issue: DLPX-61719 Reviewed-by: George Wilson <[email protected]> Reviewed-by: Paul Dagnelie <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Matthew Ahrens <[email protected]> Closes #8865
Diffstat (limited to 'cmd/ztest')
-rw-r--r--cmd/ztest/ztest.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/cmd/ztest/ztest.c b/cmd/ztest/ztest.c
index 9c2cf9501..22cfa2fe5 100644
--- a/cmd/ztest/ztest.c
+++ b/cmd/ztest/ztest.c
@@ -1692,7 +1692,7 @@ ztest_log_write(ztest_ds_t *zd, dmu_tx_t *tx, lr_write_t *lr)
if (zil_replaying(zd->zd_zilog, tx))
return;
- if (lr->lr_length > ZIL_MAX_LOG_DATA)
+ if (lr->lr_length > zil_max_log_data(zd->zd_zilog))
write_state = WR_INDIRECT;
itx = zil_itx_create(TX_WRITE,