Back to bug 1063229
| Who | When | What | Removed | Added |
|---|---|---|---|---|
| Venky Shankar | 2014-02-11 11:33:13 UTC | Status | NEW | ASSIGNED |
| CC | vshankar | |||
| Assignee | rhs-bugs | vshankar | ||
| Venky Shankar | 2014-02-13 11:45:50 UTC | Assignee | vshankar | avishwan |
| Avra Sengupta | 2014-02-13 14:00:46 UTC | CC | asengupt | |
| M S Vishwanath Bhat | 2014-02-17 06:39:08 UTC | Summary | dist-geo-rep: Few regular files are not synced to slave after geo-rep upgrade from 44rhs version to 59rhs | dist-geo-rep: Few regular files are not synced to slave when node is taken down updated and then brought back online |
| Gowrishankar Rajaiyan | 2014-02-17 06:40:05 UTC | Blocks | 1035040 | |
| Aravinda VK | 2014-02-19 10:54:11 UTC | Doc Text | Cause: After upgrade two geo-rep monitor process was running for same session. Both process were trying to use the same xsync changelog file to record the changes. Consequence: If more than one worker process uses same xsync changelog file to record changes, later one will overwrite the earlier recorded changes. These changelog files will have the list of changes to be synced to slave. Because of overwrite some files entries will miss in changelog file, and those files will not be synced to slave. Workaround (if any): before running 'geo-rep create force' kill the geo-rep monitor process. Result: When we cleanup geo-rep residue processes and start again, only one geo-rep monitor will run. | |
| Doc Type | Bug Fix | Known Issue | ||
| Shalaka | 2014-02-24 05:28:31 UTC | CC | sharne | |
| Doc Text | Cause: After upgrade two geo-rep monitor process was running for same session. Both process were trying to use the same xsync changelog file to record the changes. Consequence: If more than one worker process uses same xsync changelog file to record changes, later one will overwrite the earlier recorded changes. These changelog files will have the list of changes to be synced to slave. Because of overwrite some files entries will miss in changelog file, and those files will not be synced to slave. Workaround (if any): before running 'geo-rep create force' kill the geo-rep monitor process. Result: When we cleanup geo-rep residue processes and start again, only one geo-rep monitor will run. | After upgrade, two geo-rep monitor processes were running for same session. Both process were trying to use the same xsync changelog file to record the changes. Workaround: Before running 'geo-rep create force' command, kill the geo-rep monitor process. |
||
| Sudhir D | 2014-08-22 09:21:48 UTC | QA Contact | sdharane | storage-qa-internal |
| Aravinda VK | 2014-12-24 10:54:56 UTC | Priority | unspecified | high |
| Aravinda VK | 2015-03-06 10:57:25 UTC | Whiteboard | consistency | |
| Nagaprasad Sathyanarayana | 2015-03-17 14:36:27 UTC | Blocks | 1202842 | |
| Nagaprasad Sathyanarayana | 2015-03-26 14:48:08 UTC | Target Milestone | --- | RHS 3.1.0 |
| Aravinda VK | 2015-03-27 17:07:42 UTC | Status | ASSIGNED | CLOSED |
| Resolution | --- | DUPLICATE | ||
| Target Milestone | RHS 3.1.0 | --- | ||
| Last Closed | 2015-03-27 13:07:42 UTC | |||
| Vivek Agarwal | 2015-04-16 14:06:15 UTC | Blocks | 1202842 | |
| John Skeoch | 2016-06-01 01:56:38 UTC | CC | mzywusko |
Back to bug 1063229