A log wrap gets detected on a multi node cluster with the following error messages:
EVENT LOG:
07/05/08 15:59 Process exited sp_ocap (for o.splex) [pid = 29841] - exit(1)
07/05/08 15:59 Warning: 10727 - Possible log wrap at activation 36, log sequence number 0, offset 0 [sp_ocap/29841]
07/05/08 15:59 Warning: 10751 - log wrap detected [sp_ocap/29841]
07/05/08 15:59 Notice: Begin record skip at seqno=417963 offset=303372568 [sp_ocap/29841]
07/05/08 15:59 Warning: A portion of the redo log could not be parsed [sp_ocap/29841]
07/05/08 15:59 Notice: Using archive log /archive/log/dest/ora8_2_142102.arc [sp_ocap/29841]
The *ocap* log or event_log states the following :
09/21/02 00:15 Notice: sp_ocap (for o.splex) Failed to open "/archive/log/dest/node2" to read archive
09/21/02 00:15 Notice: sp_ocap (for o.splex) Failed to open "/archive/log/dest/node2" to read archive
Shareplex is unable to access the directory to read the archived redo logs.
1. Verify that the directory is set correctly by using the SP_OCT_ARCH_LOC parameter.
sp_ctrl>set param SP_OCT_ARCH_LOC "!;node1-sid;/archive/log/dest/node1;node2-sid;/archive/log/dest/node2"
2. If the setting is correct and the archived redo logs do exists in the location, then make sure that the archived redo logs are on a shared drive. Each node must be able to access the other node's archived redo logs. If the nodes are not on a shared drive, then you will need to copy the archived redo logs from the non-accessible location to the accessible location. You will also need to change the SP_OCT_ARCH_LOC to reflect the absolute path of the non-accessible node.For example, if you cannot access the archive log destination path of node 2, then copy the archived redo logs from node 2 and put them on the archive log destination of path 1 and update the SP_OCT_ARCH_LOC from: SP_OCT_ARCH_LOC "!;node1-sid;/archive/log/dest/node1;node2-sid;/archive/log/dest/node2" to SP_OCT_ARCH_LOC "!;node1-sid;/archive/log/dest/node1;node2-sid;/archive/log/dest/node1".
This solution assumes that the all the nodes are open. You can run the following query to verify the information:
select thread#, instance_name, status from gv$instance;
---------------------------------------- In later versions like SharePlex 9.1 and up, the event log messages are modified as shown below: Info 2018-03-27 16:51:44.604251 29813 643084816 Capture exited with code=1, pid = 29814 (capturing from o.ORA11GR2) Error 2018-03-27 16:51:44.601373 29814 1114429760 Capture: [SP-OCT01001] Unable to find archivelog with sequence#: 77041 (thread: 1).
