[MGNLEE-645] Cannot stop Magnolia since tomcat upgrade Created: 17/Jun/21 Updated: 26/Apr/22 Resolved: 25/Mar/22 |
|
| Status: | Closed |
| Project: | Magnolia DX Core |
| Component/s: | None |
| Affects Version/s: | 6.2.7, 6.2.15, 6.2.17 |
| Fix Version/s: | 6.2.18 |
| Type: | Bug | Priority: | Neutral |
| Reporter: | Richard Gange | Assignee: | Michael Duerig |
| Resolution: | Fixed | Votes: | 1 |
| Labels: | foundation_team, maintenance | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Attachments: |
|
||||||||||||||||||||||||||||||||||||
| Issue Links: |
|
||||||||||||||||||||||||||||||||||||
| Template: |
|
||||||||||||||||||||||||||||||||||||
| Acceptance criteria: |
Empty
|
||||||||||||||||||||||||||||||||||||
| Task DoD: |
[X]*
Doc/release notes changes? Comment present?
[X]*
Downstream builds green?
[X]*
Solution information and context easily available?
[X]*
Tests
[X]*
FixVersion filled and not yet released
[ ] 
Architecture Decision Record (ADR)
|
||||||||||||||||||||||||||||||||||||
| Bug DoR: |
[X]*
Steps to reproduce, expected, and actual results filled
[X]*
Affected version filled
|
||||||||||||||||||||||||||||||||||||
| Date of First Response: | |||||||||||||||||||||||||||||||||||||
| Description |
|
Since the upgrade |
| Comments |
| Comment by Christopher Chard [ 01/Feb/22 ] |
|
And I thought it was only me... happens to us as well |
| Comment by Maxime Michel [ 28/Feb/22 ] |
|
FTR mduerig is reporting having seen this issue in his IDE with an old Tomcat version (9.0.10). This makes us doubt this being from the Tomcat upgrade. Also, additional debugging information can be found in the attached thread dump. dai.ha any reason you never submitted your patch as a PR? |
| Comment by Maxime Michel [ 28/Feb/22 ] |
|
TODO: move this ticket to the project where the actual issue is coming from. (UI?) |
| Comment by Richard Gange [ 28/Feb/22 ] |
|
Sounds good. I wasn't entirely sure it was tomcat. Looking back through the changelog |
| Comment by Dai Ha [ 02/Mar/22 ] |
|
mmichel, at that time I was just merely fix for local setup, and these investigations are for preference only and I was take care other topics. For now, what I can say is the problems are still there till now, as mentioned previously, the branch on ui helps most in my case. |
| Comment by Michael Duerig [ 17/Mar/22 ] |
|
Re. the h2 driver leak: org.apache.catalina.loader.WebappClassLoaderBase.clearReferencesJdbc The web application [magnoliaAuthor] registered the JDBC driver [org.h2.Driver] but failed to unregister it when the web application was stopped. To prevent a memory leak, the JDBC Driver has been forcibly unregistered. The h2 driver is registered automatically as part of Java's service provider mechanism and is apparently not unregistered on shutdown. It is unclear to me whose responsibility this would even be and whether the service provider mechanism even support this. In any case the impact of not deregistering the driver is minimal as there is only ever one static instance being used. We should not deregister the driver along with the persistent manager (as proposed in this patch), as it might still be used from an other persistent manager. Best solution IMO is to do nothing (apart from maybe documenting as know issue) and rely on Tomcat doing the right thing. /cc dai.ha |
| Comment by Michael Duerig [ 25/Mar/22 ] |
|
I could not reproduce the thread local leak. Filed |