[LU-1811] In replay-barrier, it should sync all of operations before setting rdonly of the dev. Created: 31/Aug/12  Updated: 31/Aug/12

Status: Open
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Minor
Reporter: Di Wang Assignee: Di Wang
Resolution: Unresolved Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 9762

 Description   

In replay-barrier

replay_barrier() {
local facet=$1
do_facet $facet "sync; sync; sync"
df $MOUNT

  1. make sure there will be no seq change
    local clients=${CLIENTS:-$HOSTNAME}
    do_nodes $clients "f=${MOUNT}/fsa-\\\$(hostname); mcreate \\\$f; rm \\\$f"

local svc=${facet}_svc
do_facet $facet $LCTL --device %${!svc} notransno
do_facet $facet $LCTL --device %${!svc} readonly
do_facet $facet $LCTL mark "$facet REPLAY BARRIER on ${!svc}"
$LCTL mark "local REPLAY BARRIER on ${!svc}"
}
"sync sync sync" should be moved after the mcreate, so all of the ops should be committed before the barrier, otherwise mcreate might be replayed as well, which sometimes confuse the test results.



 Comments   
Comment by Di Wang [ 31/Aug/12 ]

http://review.whamcloud.com/3842

Generated at Sat Feb 10 01:19:53 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.