During the execution of publish-dist (part of the do-release.sh), I found the strange problem with commit 748ec455b33d29ed007b0143efa73b7409248da7 from HBASE-27498, the test in hbase-it/src/test/java/org/apache/hadoop/hbase/client/TestConnectionImplementation.java is blocking the javadoc target test-aggregate-no-fork when running mvn -B --settings <local repo setting.xml> site -DskipTests with the following ERROR.
I tried to revert 748ec455b33d29ed007b0143efa73b7409248da7 or revert just the TestConnectionImplementation class, then the javadoc/site completes without error. I also tried to checkout just branch-2.5 and branch-2.4, execute the build and javadoc/site execution, but it is still failing. # capture using JDK17 , it has the same NullPointerException exception for JDK8 and JDK11 without much detail. error: An internal exception has occurred. (java.lang.NullPointerException: Cannot invoke "com.sun.source.util.TreePath.getCompilationUnit()" because the return value of "com.sun.source.util.DocTrees.getPath(javax.lang.model.element.Element)" is null) Please file a bug against the javadoc tool via the Java bug reporting page (https://bugreport.java.com) after checking the Bug Database (https://bugs.java.com) for duplicates. Include error messages and the following diagnostic in your report. Thank you. java.lang.NullPointerException: Cannot invoke "com.sun.source.util.TreePath.getCompilationUnit()" because the return value of "com.sun.source.util.DocTrees.getPath(javax.lang.model.element.Element)" is null at jdk.javadoc/jdk.javadoc.internal.doclets.toolkit.util.Utils.getFileObject(Utils.java:871) at jdk.javadoc/jdk.javadoc.internal.doclets.formats.html.SourceToHTMLConverter.convertClass(SourceToHTMLConverter.java:198) at jdk.javadoc/jdk.javadoc.internal.doclets.formats.html.SourceToHTMLConverter.generate(SourceToHTMLConverter.java:137) at jdk.javadoc/jdk.javadoc.internal.doclets.formats.html.SourceToHTMLConverter.convertRoot(SourceToHTMLConverter.java:112) at jdk.javadoc/jdk.javadoc.internal.doclets.formats.html.HtmlDoclet.generateOtherFiles(HtmlDoclet.java:218) at jdk.javadoc/jdk.javadoc.internal.doclets.toolkit.AbstractDoclet.startGeneration(AbstractDoclet.java:218) at jdk.javadoc/jdk.javadoc.internal.doclets.toolkit.AbstractDoclet.run(AbstractDoclet.java:115) at jdk.javadoc/jdk.javadoc.doclet.StandardDoclet.run(StandardDoclet.java:103) at jdk.javadoc/jdk.javadoc.internal.tool.Start.parseAndExecute(Start.java:556) at jdk.javadoc/jdk.javadoc.internal.tool.Start.begin(Start.java:393) at jdk.javadoc/jdk.javadoc.internal.tool.Start.begin(Start.java:342) at jdk.javadoc/jdk.javadoc.internal.tool.Main.execute(Main.java:63) at jdk.javadoc/jdk.javadoc.internal.tool.Main.main(Main.java:52) 1 error 100 warnings May I ask if you checkout the branch-2.5 and run the following to see if I'm the only one hitting the error ? the command is mvn clean install -DskipTests && mvn -B site -DskipTests Secondly, other than reverting this change, do you have any suggestion how we can debug javadoc generation (I tried javadoc -J-Xmx2G @options @argfile -verbose, but this didn't tell me anything...) -Stephen On Mon, Jan 23, 2023 at 9:01 PM Tak Lon (Stephen) Wu <[email protected]> wrote: > > Fyi I’m executing the release build via do-release-docker.sh, ran till the > mvn site and Java doc generation. > > But it failed with 20:05:46 [ERROR] javadoc: error - > java.lang.NullPointerException , and few error/warning for @result tag as > well as @link not found for the generated proto class BlockingInterface. > > I’m still working on it, if possible, please avoid to push commits to > branch-2.5 > > Thanks, > Stephen > > On Sun, Jan 22, 2023 at 1:29 PM Andrew Purtell <[email protected]> wrote: >> >> Sounds good Stephen, please ping me as you need assistance. >> >> On Sat, Jan 21, 2023 at 9:32 PM Tak Lon (Stephen) Wu <[email protected]> >> wrote: >> >> > Thanks Andrew and Bryan, all the blockers for branch-2.5 should have >> > been resolved. >> > >> > I'm running the compatibility check again, and will continue the >> > release tasks tomorrow or Monday morning (PST). >> > >> > -Stephen >> > >> > On Fri, Jan 20, 2023 at 11:16 AM Bryan Beaudreault >> > <[email protected]> wrote: >> > > >> > > You probably saw, but just to close the loop here -- HBASE-27579 got >> > > resolved, so no blocker there anymore. Thanks all! >> > > >> > > On Fri, Jan 20, 2023 at 12:30 PM Andrew Purtell <[email protected]> >> > wrote: >> > > >> > > > I think HBASE-27539 should be reverted from the releasing branches. >> > Keeping >> > > > it in branch-2 and master would be fine, we do frequently change LP >> > > > interface details in minor releases. It is the compatibility break in a >> > > > patch release with a real affected downstream (Phoenix) that is a >> > problem. >> > > > As RM I would go ahead and do that, I suggest the same to you. >> > > > >> > > > On Thu, Jan 19, 2023 at 10:25 PM Tak Lon (Stephen) Wu < >> > [email protected]> >> > > > wrote: >> > > > >> > > > > Thanks Andrew for sharing the normal flow and helped me off this >> > > > > thread. I have tried steps 1 and 2, and found the following JIRAs may >> > > > > need a few days before closing for 2.5.3 RC. >> > > > > >> > > > > HBASE-27579 under review >> > > > > HBASE-27539 broke the patch version backward compatibility with >> > > > > StoreFileReader, pending confirmation if we can revert >> > > > > HBASE-27578 should be merging soon >> > > > > >> > > > > -Stephen >> > > > > >> > > > > On Thu, Jan 19, 2023 at 8:34 AM Andrew Purtell < >> > [email protected] >> > > > > >> > > > > wrote: >> > > > > > >> > > > > > Sure. We are not rushing. >> > > > > > >> > > > > > Welcome to the RM club. You have commit privileges so all steps >> > > > outlined >> > > > > below should be achievable. Let me briefly outline the steps for >> > making a >> > > > > release. We can deal with details privately by email. >> > > > > > >> > > > > > 1. Run the compatibility checker. Verify no changes from prior >> > release >> > > > > that are not allowed. Exceptions are fine if you as RM are willing to >> > > > > defend them when the PMC has questions. If there are problems that >> > need >> > > > > fixing, fix them. >> > > > > > >> > > > > > 2. Clean up fix versions in JIRA. All issues targeting the release >> > eg >> > > > > 2.5.3 should be resolved with that fix version in the set if a >> > commit was >> > > > > made, and if no commit but still resolved the fix version should be >> > > > removed >> > > > > from the set. If the issue is not resolved the fix version should be >> > > > > removed and replaced with one for the next eg 2.5.3 -> 2.5.4. When >> > you >> > > > are >> > > > > done a report for the fix version of your release candidate should >> > show >> > > > all >> > > > > issues resolved. >> > > > > > >> > > > > > 3. Run the create-release script. It will handle the production and >> > > > > staging of artifacts end to end. This includes the production of >> > release >> > > > > notes derived from JIRA. This includes tagging in git on the branch. >> > The >> > > > > script will ask for your Apache committer credentials at the start. >> > > > > > >> > > > > > 4. Take the artifacts produced by create-release and sanity check >> > them. >> > > > > Check sums and signatures. Unpack and check included contents. >> > Launch a >> > > > > minicluster, load some data with PE or LTT. The same actions you’d >> > take >> > > > if >> > > > > voting on someone else’s release candidate. Assuming no issue, >> > proceed. >> > > > > > >> > > > > > 5. Send the vote mail produced by create-release to get the vote >> > > > started. >> > > > > > >> > > > > > 6. Respond to comments on the vote thread as required. >> > > > > > >> > > > > > 7. Close the vote once sufficient votes have been received. Give >> > it a >> > > > > week at least. If it seems like voting is slow to complete gently >> > prod >> > > > the >> > > > > PMC on dev@. >> > > > > > >> > > > > > 8. Create and push a release tag in git. Move the release candidate >> > > > from >> > > > > staging to release in svn. >> > > > > > >> > > > > > 9. Edit download.xml and trigger a new site build. Wait for your >> > > > changes >> > > > > to go live. >> > > > > > >> > > > > > 10. Send out a release announcement email. >> > > > > > >> > > > > > > On Jan 18, 2023, at 8:49 PM, Tak Lon (Stephen) Wu < >> > [email protected] >> > > > > >> > > > > wrote: >> > > > > > > >> > > > > > > Hi Andrew, >> > > > > > > >> > > > > > > If we’re not rushing, I would like to help release or at least >> > learn >> > > > > as a >> > > > > > > shadow release manager how to do it end-to-end. >> > > > > > > >> > > > > > > I will ping you separately. >> > > > > > > >> > > > > > > Thanks, >> > > > > > > Stephen >> > > > > > > >> > > > > > >> On Wed, Jan 18, 2023 at 8:35 PM Andrew Purtell < >> > > > > [email protected]> >> > > > > > >> wrote: >> > > > > > >> >> > > > > > >> Ok. >> > > > > > >> >> > > > > > >>> On Jan 18, 2023, at 8:28 PM, Bryan Beaudreault < >> > > > > [email protected]> >> > > > > > >> wrote: >> > > > > > >>> >> > > > > > >>> I just submitted a blocker jira: >> > > > > > >>> https://issues.apache.org/jira/browse/HBASE-27579 >> > > > <https://issues.apache.org/jira/browse/HBASE-27579> >> > > > > > >>> >> > > > > > >>> This issue is old, not new to 2.4.x or 2.5.x. But I think we >> > should >> > > > > try >> > > > > > >> to >> > > > > > >>> get the fix into the next patch releases, since it is data loss >> > > > > related. >> > > > > > >>> >> > > > > > >>>> On Tue, Jan 17, 2023 at 5:19 PM Andrew Purtell < >> > > > [email protected] >> > > > > > >> > > > > > >> wrote: >> > > > > > >>>> >> > > > > > >>>> I've cleaned up JIRA for fix versions 2.4.16 and 2.5.3. There >> > is a >> > > > > > >>>> dependabot PR to fix a warning, see HBASE-27575. When this is >> > > > > merged, I >> > > > > > >>>> assume tomorrow, I'll go ahead and prepare the two releases, >> > > > unless >> > > > > Nick >> > > > > > >>>> you want to pick one. Just say which. >> > > > > > >>>> >> > > > > > >>>> >> > > > > > >>>> On Tue, Jan 17, 2023 at 1:07 AM 张铎(Duo Zhang) < >> > > > > [email protected]> >> > > > > > >>>> wrote: >> > > > > > >>>> >> > > > > > >>>>> Good. Thanks Nick and Andrew. Since we have two releases, I >> > think >> > > > > each >> > > > > > >>>>> of you can pick one branch to release :) >> > > > > > >>>>> >> > > > > > >>>>> Anyway, let me make a new thirdparty release and bump the >> > > > > dependencies >> > > > > > >>>>> first, IIRC there is at least a CVE fix in netty. >> > > > > > >>>>> >> > > > > > >>>>> Thanks. >> > > > > > >>>>> >> > > > > > >>>>> Andrew Purtell <[email protected]> 于2023年1月17日周二 >> > 03:13写道： >> > > > > > >>>>>> >> > > > > > >>>>>> I’ll be happy to run a release or both of them as needed >> > too. >> > > > > > >>>>>> >> > > > > > >>>>>>> On Jan 16, 2023, at 6:44 AM, Nick Dimiduk < >> > [email protected] >> > > > > >> > > > > > >>>> wrote: >> > > > > > >>>>>>> >> > > > > > >>>>>>> On Mon, Jan 16, 2023 at 15:02 张铎(Duo Zhang) < >> > > > > [email protected]> >> > > > > > >>>>> wrote: >> > > > > > >>>>>>> >> > > > > > >>>>>>>> If the release managers do not have enough time, do we >> > have >> > > > any >> > > > > > >>>>> volunteers? >> > > > > > >>>>>>> >> > > > > > >>>>>>> >> > > > > > >>>>>>> I’m happy to help out as needed. >> > > > > > >>>>>>> >> > > > > > >>>>>>> Thanks, >> > > > > > >>>>>>> Nick >> > > > > > >>>>>>> >> > > > > > >>>>>>>> >> > > > > > >>>>> >> > > > > > >>>> >> > > > > > >>>> >> > > > > > >>>> -- >> > > > > > >>>> Best regards, >> > > > > > >>>> Andrew >> > > > > > >>>> >> > > > > > >>>> Unrest, ignorance distilled, nihilistic imbeciles - >> > > > > > >>>> It's what we’ve earned >> > > > > > >>>> Welcome, apocalypse, what’s taken you so long? >> > > > > > >>>> Bring us the fitting end that we’ve been counting on >> > > > > > >>>> - A23, Welcome, Apocalypse >> > > > > > >>>> >> > > > > > >> >> > > > > >> > > > >> > > > >> > > > -- >> > > > Best regards, >> > > > Andrew >> > > > >> > > > Unrest, ignorance distilled, nihilistic imbeciles - >> > > > It's what we’ve earned >> > > > Welcome, apocalypse, what’s taken you so long? >> > > > Bring us the fitting end that we’ve been counting on >> > > > - A23, Welcome, Apocalypse >> > > > >> > >> >> >> -- >> Best regards, >> Andrew >> >> Unrest, ignorance distilled, nihilistic imbeciles - >> It's what we’ve earned >> Welcome, apocalypse, what’s taken you so long? >> Bring us the fitting end that we’ve been counting on >> - A23, Welcome, Apocalypse > > -- > Sent from Gmail Mobile
