Re: [PR] Sessioncounter [tomcat]

2024-04-12 Thread via GitHub
rmaucher commented on PR #713: URL: https://github.com/apache/tomcat/pull/713#issuecomment-2051302240 Thanks for pointing this out. Since tracking too many redundant stats is not super useful, I decided to remove this one (it is now approximated by adding the active sessions to the expired

Re: [PR] Sessioncounter [tomcat]

2024-04-12 Thread via GitHub
rmaucher closed pull request #713: Sessioncounter URL: https://github.com/apache/tomcat/pull/713 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

Re: [PR] Sessioncounter [tomcat]

2024-04-08 Thread via GitHub
rmaucher commented on PR #713: URL: https://github.com/apache/tomcat/pull/713#issuecomment-2042934315 I think this field should simply be dropped in Tomcat 11, it's redundant with the others. Shouldn't it be expired + active ? For the other branches I would deprecate the field. -- This

Re: [PR] Sessioncounter [tomcat]

2024-04-07 Thread via GitHub
pangxianhai commented on PR #713: URL: https://github.com/apache/tomcat/pull/713#issuecomment-2041713809 > Is this fixing an observed issue or a theoretical one? observed issue. On the monitoring platform, we found that the sessionCounter is less than the expiredSessions. The

Re: [PR] Sessioncounter [tomcat]

2024-04-07 Thread via GitHub
markt-asf commented on PR #713: URL: https://github.com/apache/tomcat/pull/713#issuecomment-2041577898 Is this fixing an observed issue or a theoretical one? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[PR] Sessioncounter [tomcat]

2024-04-07 Thread via GitHub
pangxianhai opened a new pull request, #713: URL: https://github.com/apache/tomcat/pull/713 sessionCounter change to AtomicLong. Avoid inaccurate counting in concurrent situations -- This is an automated message from the Apache Git Service. To respond to the message, please log on to