| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| ... | |
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
changes.
Multi-value'ed attribute diff was not implemented; Added the implementation.
|
| |
|
|
| |
(checked in on behalf of dboreham)
|
| | |
|
| |
|
|
| |
Convert to C-style comments to avoid build breakage.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
description
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
dblayer_close
1) Let dblayer_force_checkpoint check if the db env is still good or not.
If it's already released, it just returns.
2) In _dblayer_delete_instance_dir, it actually does not need to invoke db
threads. Set DBLAYER_NO_DBTHREADS_MODE when calling dblayer_start.
|
| | |
|
| |
|
|
| |
reversed the logic to set change_is_relevant, which triggers to send notify.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Bug Description: performance counters gone after replication initialization
Reviewed by: Nathan, Noriko, and David (Thanks!)
Fix Description: There is a state flag in the backend object. When repl init starts, the bulk import start code eventually calls dblayer_instance_close which does some closure of the backend, including setting the state flag to STOPPED. However, when repl init completes, the bulk import code eventually calls dblayer_instance_start which does not set the state flag. The only place where the state flag is set is in ldbm_instance_start (which also calls dblayer_instance_start). From my brief investigation, it doesn't appear that ldbm_instance_stop and start are all that useful - they only deal with changing the state flag to stopped or started, and that is used in very few places, the monitor code being one of them. This fix makes it "symmetrical" e.g. you undo in dblayer_instance_start what was done in dblayer_instance_close.
A better solution might be to call ldbm_instance_stop instead of dblayer_instance_close from bulk_import_start and remove the code to set the state to stopped from dblayer_instance_close. Then do the same in import_all_done - call ldbm_instance_start instead of dblayer_instance_start. The only difference is the locking for be_state_lock - perhaps some programmer decided to use the dblayer function instead of the ldbm function to avoid deadlock? But I don't think doing that would introduce any more chances of deadlock. The other problem with this is that ldbm_instance_stop deletes the cache, while dblayer_instance_close merely cleans it out.
Platforms tested: RHEL3
Flag Day: no
Doc impact: no
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
The width of time in changelog is always 4 bytes.
|
| |
|
|
|
|
| |
file, -c -- changelog file) for any type of files, which could sometimes crashes the command. To prevent it, introduced a simple file type checker and reset the file type if it does not match the real type.
2) key search was not working when the key has multiple values.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Bug Description: customer requests that the server warns that the files are missing (the current errors are generic about being unable to retrieve server-cert)
Reviewed by: Nathan and Noriko (Thanks!)
Files: ssl.c
Branch: HEAD
Fix Description: Basically just check to see if the specified files are readable. If not, check for newer versions e.g. if the customer hasn't changed the cert filename from *-cert7.db to *-cert8.db yet. Also threw in key4.db in case that changes in the future (not likely). Log a warning if the files do not exist (most likely the server will fail to start in this case).
Platforms tested: RHEL3
Flag Day: no
Doc impact: no, I don't think so
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| | |
|
| |
|
|
|
|
|
|
| |
options
1) if "max length of dumpled id list" is larger than 1 MB, it reduces it to 1 MB.
2) if "max length of dumpled id list" is less than 40B, it increases it to 40B.
3) if the idl is not dumpled in the MAX_BUFFER size buffer, it repeats till the dump is done.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
| |
Bug Description: Many operations use write_audit_log_entry to write to the audit log. These functions build a string buffer to write to the audit log. The only problem is, they never check to see if audit logging is enabled until after all of this work has been done. write_audit_log_entry should check right away to see if logging is enabled. Should save some time for write operations.
Reviewed by: Noriko (Thanks!)
Fix Description: The code in auditlog.c is called for every update operation, and in many cases a lot of work is done before it is even known if audit logging is enabled. Perhaps this was because there was no function that could be called outside of the logging code to see if audit logging is enabled. I added config_get_auditlog_logging_enabled() and used it in auditlog.c to skip the code altogether if audit logging is disabled (the default).
Platforms tested: RHEL3
Flag Day: no
Doc impact: no, the functions are in the private API
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
