Date
1 - 9 of 9
topics for integration meeting tomorrow (12/11/2019)?
|
JamO Luhrsen
Please write back if you have any agenda item for the meeting
window we have tomorrow 9pm PST. If not much, we will cancel. Thanks, JamO |
|
|
|
Luis Gomez
No agenda item but FYI it is very possible we will have to disable JVM monitoring by default in all jobs. Reason being is new decanter version we are using to fix issues in master is not stable in the cluster jobs (member delayed startup, sometimes the boot process is not completed, etc). I am investigating some workarounds but so far no luck :(
toggle quoted message
Show quoted text
BR/Luis On Dec 10, 2019, at 2:56 PM, JamO Luhrsen <jluhrsen@...> wrote: |
|
|
|
I'd like to review the possibility of picking up a build for neon sr3 as well as the code freeze for service releases thanks On Tue, Dec 10, 2019 at 2:56 PM JamO Luhrsen <jluhrsen@...> wrote: Please write back if you have any agenda item for the meeting --
|
|
|
|
JamO Luhrsen
I just realized I cannot attend the 9pm meeting tomorrow.
toggle quoted message
Show quoted text
Daniel, do you want to try to carve some time out tomorrow earlier in the day to discuss the release topics. Seems that it's our only topic at this point anyway. JamO On 12/10/19 5:07 PM, Daniel De La Rosa
wrote:
|
|
|
|
Sure, that works for me. I’ll look for a time slot On Tue, Dec 10, 2019 at 11:22 PM Jamo Luhrsen <jluhrsen@...> wrote:
--
|
|
|
|
Robert Varga
On 11/12/2019 00:44, Luis Gomez wrote:
Reason being is new decanter version we are using to fix issues in master is not stable in the cluster jobs (member delayed startup, sometimes the boot process is not completed, etc). I am investigating some workarounds but so far no luck :(Hmm... we do have the odlparent-6.0.3 lot coming up, which upgrades karaf from 4.2.2 to 4.2.6, if that's any help. How does latest decanter work on stable/neon? (that already has karaf-4.2.6). Regards, Robert |
|
|
|
Luis Gomez
The decanter situation is what I said in the other mail:
toggle quoted message
Show quoted text
- Old decanter version 1.0.X: works well in neon and sodium but not in master so we have to skip magnesium and any future release for now. - Updated decanter version 1.X: works in all branches but cluster start is impacted so we have to skip all cluster tests. - Latest decanter version 2.X: does not work in any branch, it could be some configuration/ES integration problem, I guess we will need to spend some cycles here to make it work. BR/Luis On Dec 11, 2019, at 12:09 PM, Robert Varga <nite@...> wrote: |
|
|
|
Luis Gomez
BTW for the first issue (old version not working in Magnesium) I suspect it is JDK 11 problem.
toggle quoted message
Show quoted text
You can can see the ERRORs in this log: https://logs.opendaylight.org/sandbox/vex-yul-odl-jenkins-2/openflowplugin-csit-1node-perf-bulkomatic-only-magnesium/5/odl_1/odl1_karaf.log.gz BR/Luis
|
|
|
|
Robert Varga
On 11/12/2019 22:50, Luis Gomez wrote:
BTW for the first issue (old version not working in Magnesium) I suspectYeah: Caused by: java.lang.UnsupportedOperationException: Boot class path mechanism is not supportedhttps://www.oracle.com/technetwork/java/javase/9-relnote-issues-3704069.html : The boot class path has been mostly removed in this release. The java -Xbootclasspath and -Xbootclasspath/p options have been removed. The javac -bootclaspath option can only be used when compiling to JDK 8 or older. The system property sun.boot.class.path has been removed. Deployments that rely on overriding platform classes for testing purposes with -Xbootclasspath/p will need to changed to use the --patch-module option that is documented in JEP 261. The -Xbootclasspath/a option is unchanged.I think this is a bug in ES, which fails to account for UOE being legally thrown from RuntimeMXBean.getBootClassPath(). https://stackoverflow.com/questions/56167397/elasticsearch-1-6-could-not-initialize-class-org-elasticsearch-monitor-jvm-jvmin, too :) Regards, Robert |
|
|