On 12/13/2011 12:44 AM, Yunkai Zhang wrote:
> Hi Steven Dake:
>
> How do you think about this patch v2?
>
> According our analysis above, I think this patch is useful.
>
With any totem patches in a stable branch there needs to be a failure
(segfault, improper membership behavior, message loss) preferably with a
test case. I am super leary of making _any_ changes to this code
without significant motivation.
Regards
-steve
> On Mon, Dec 12, 2011 at 6:07 PM, Yunkai Zhang <qiushu.zyk@xxxxxxxxxx> wrote:
>> After the representative of the new ring entered into RECOVERY state,
>> it was waiting for the commit_token on the second rotation. But when
>> it got a new commit_token, it didn't check the status of this token
>> which could be retransmitted by upstream node on the first rotation,
>> and then the representative node would send the initial ORF token
>> prematurely.
>>
>> Signed-off-by: Yunkai Zhang <qiushu.zyk@xxxxxxxxxx>
>> ---
>> exec/totemsrp.c | 4 +++-
>> 1 files changed, 3 insertions(+), 1 deletions(-)
>>
>> diff --git a/exec/totemsrp.c b/exec/totemsrp.c
>> index 5a78962..7a4be95 100644
>> --- a/exec/totemsrp.c
>> +++ b/exec/totemsrp.c
>> @@ -4372,7 +4372,9 @@ static int message_handler_memb_commit_token (
>> break;
>>
>> case MEMB_STATE_RECOVERY:
>> - if (totemip_equal (&instance->my_id.addr[0], &instance->my_ring_id.rep)) {
>> + if (totemip_equal (&instance->my_id.addr[0], &instance->my_ring_id.rep) &&
>> + memb_commit_token->ring_id.seq == instance->my_ring_id.seq &&
>> + memb_commit_token->token_seq == 2*memb_commit_token->addr_entries) {
>> log_printf (instance->totemsrp_log_level_debug,
>> "Sending initial ORF token\n");
>>
>> --
>> 1.7.7.3
>>
>
>
>
_______________________________________________
discuss mailing list
discuss@xxxxxxxxxxxx
http://lists.corosync.org/mailman/listinfo/discuss
