|
Mantis - Resin
|
|||||
| Viewing Issue Advanced Details | |||||
|
|
|||||
| ID: | Category: | Severity: | Reproducibility: | Date Submitted: | Last Update: |
| 219 | minor | always | 05-26-05 00:00 | 11-30-05 14:43 | |
|
|
|||||
| Reporter: | user202 | Platform: | |||
| Assigned To: | OS: | ||||
| Priority: | urgent | OS Version: | |||
| Status: | closed | Product Version: | 3.0.13 | ||
| Product Build: | 3.0.13 | Resolution: | duplicate | ||
| Projection: | none | ||||
| ETA: | none | Fixed in Version: | |||
|
|
|||||
| Summary: | 0000219: mod_caucho Breaks After 3 Apache Restarts | ||||
| Description: |
RSN-242 The following procedure consistently causes 503 errors when accessing a Resin instance through a 3.0.13 Resin version of mod_caucho: - Start the Resin container with the default ROOT webapp deployed - Hit the webapp through Apache/mod_caucho from a browser: Returns "Directory of /" - Restart Apache using "apachectl graceful" - Refresh the browser: Returns "Directory of /" - Restart Apache using "apachectl graceful" - Refresh the browser: Returns "Directory of /" - Restart Apache using "apachectl graceful" - Refresh the browser: Returns 503 Error Once this happens, the only way to get it working again is by doing the following: - Remove the /tmp/*_6802 files - Perform a "apachectl graceful" (or restart Apache in some other way) The apachectl graceful happens nightly on our servers as part of our Apache access log rotation. As a workaround for the time being, we have added deletion of /tmp/*_6802 files right before the "apachectl graceful" in the nightly cron job. Note that these particular steps only cause the error when using a 3.0.13 version of mod_caucho. Use of a 3.0.12 mod_caucho does not seem to generate the error after the 3rd restart. |
||||
| Steps To Reproduce: | |||||
| Additional Information: |
Red Hat Enterprise v.3 Apache 2.0.46-46.ent Sun JDK 1.4.2_07-b05 Resin 3.0.13 (OS or Pro) |
||||
| Relationships | |||||
| Attached Files: | |||||
| Notes | |||||
|
|
|||||
|
|
||||
|
|
|||||
|
|
||||