diff options
Diffstat (limited to 'issues/gn-uploader')
| -rw-r--r-- | issues/gn-uploader/build-uploader-container-20240321.org | 210 |
1 files changed, 210 insertions, 0 deletions
diff --git a/issues/gn-uploader/build-uploader-container-20240321.org b/issues/gn-uploader/build-uploader-container-20240321.org index 3511e23..61df2d8 100644 --- a/issues/gn-uploader/build-uploader-container-20240321.org +++ b/issues/gn-uploader/build-uploader-container-20240321.org @@ -420,3 +420,213 @@ Start the container for now: #+begin_src shell sudo systemctl start genenetwork-uploader-container.service #+end_src + + +* Log 2024-03-22T11:24+03:00UTC + +Verify guix is clean: +#+begin_src shell + /usr/local/guix-profiles/guix-daemon/bin/guix describe +#+end_src +giving: +#+begin_example +Generation 4 Mar 21 2024 05:04:28 (current) + guix 69951a6 + repository URL: https://git.savannah.gnu.org/git/guix.git + branch: master + commit: 69951a61a1d8f1f2135ea2dc836738be282b97bc +#+end_example + +Now clone/pull all relevant repositories +#+begin_src shell + $ cd /home/fredm/gn-machines/ && git pull origin define-gn-uploader + $ cd /home/fredm/guix-bioinformatics/ && git pull origin master + $ cd /home/fredm/ && git clone https://gitlab.inria.fr/guix-hpc/guix-past +#+end_src + all those succeeded. + + Cloning =guix-forge= failed: + #+begin_src shell + $ cd /home/fredm/ && git clone https://git.systemreboot.net/guix-forge/ + Cloning into 'guix-forge'... + fatal: unable to access 'https://git.systemreboot.net/guix-forge/': server certificate verification failed. CAfile: /etc/ssl/certs/ca-certificates.crt CRLfile: none + #+end_src +looks like *tux02* does not trust the certificates from systemreboot.net + +Clone =guix-forge= with no verification + #+begin_src shell + $ cd /home/fredm/ && env GIT_SSL_NO_VERIFY=1 git clone https://git.systemreboot.net/guix-forge/ + #+end_src + + Success!!! + + Now stop uploader container: + #+begin_src shell + sudo systemctl stop genenetwork-uploader-container.service + #+end_src + +Delete existing logs: +#+begin_src shell + sudo rm -fv /export2/guix-containers/genenetwork/uploader/var/log/gunicorn-g*.log +#+end_src + +Find out the process related to the annoying log file: +#+begin_src shell + sudo lsof /export2/guix-containers/genenetwork/uploader/var/log/gunicorn-genenetwork2.log +#+end_src +and we get: +#+begin_example +COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME +shepherd 94815 root 15w REG 259,8 2322452 9830599 /export2/guix-containers/genenetwork/uploader/var/log/gunicorn-genenetwork2.log +#+end_example + +Get into the container and check the container name: +#+begin_src shell + $ sudo /usr/local/guix-profiles/guix-daemon/bin/guix container exec 94815 /run/current-system/profile/bin/bash --login + root@genenetwork /# hostname + genenetwork +#+end_src + +Aha! Looks like I might have run the build for the uploader container on *tux02* +before I had changed the hostnames and paths! + +Check PID(s) of production container: +#+begin_src shell + $ ps -u root -f --forest | grep -A4 '/usr/local/bin/genenetwork-container' +#+end_src +which gives: +#+begin_example +root 61415 1 0 Mar20 ? 00:00:00 /gnu/store/1gd9nsy4cps8fnrd1avkc9l01l7ywiai-guile-3.0.9/bin/guile --no-auto-compile /usr/local/bin/genenetwork-container +root 61436 61415 0 Mar20 ? 00:15:27 \_ /gnu/store/bhynhk0c6ssq3fqqc59fvhxjzwywsjbb-guile-3.0.9/bin/guile --no-auto-compile /gnu/store/06mz0yjkghi7r6d7lmhvv7gryipljhdd-shepherd-0.10.3/bin/shepherd --config /gnu/store/gg29j35fvsx04xc41yb3zx7zgd09519a-shepherd.conf +root 61488 61436 0 Mar20 ? 00:00:00 \_ /gnu/store/gbz5y54xi3bxc843azjsssmv6n5p8kj3-eudev-3.2.11/sbin/udevd +root 61533 61436 0 Mar20 ? 00:00:00 \_ /gnu/store/lx54pvb5523v45i6c3axzcjlvl6z18wz-guix-1.4.0-16.aeb4943/bin/guix-daemon --build-users-group guixbuild --max-silent-time 3600 --timeout 86400 --log-compression gzip --discover=no --substitute-urls https://ci.guix.gnu.org https://bordeaux.guix.gnu.org --disable-chroot +root 61567 61436 0 Mar20 ? 00:00:16 \_ /gnu/store/6i3bj0j8m97rmgdsg2vgrx38crpmnwan-inetutils-2.3/libexec/syslogd --rcfile=/etc/syslog.conf +#+end_example + +So the container that is shouting into the log file is not the production container! Awesome! We can safely kill the process. + +First off, let's try and figure out the parent PID for the process: +#+begin_src shell + ps -f --forest -p 94815 +#+end_src + +which gives: +#+begin_example +UID PID PPID C STIME TTY TIME CMD +root 94815 1 13 Mar12 pts/31 1-08:33:46 /gnu/store/bhynhk0c6ssq3fqqc59fvhxjzwywsjbb-guile-3.0.9/bin/guile --no-auto-compile /gnu/store/06m +#+end_example + +There are no other related processes! Looks like an orphaned process from a possibly older container… + +Kill it! +#+begin_src shell + sudo kill -s SIGKILL 94815 +#+end_src + +Check whether production (test1.genenetwork.org) is still online +#+begin_src + systemctl status genenetwork-container.service +#+end_src + +Yep! We are good! + +Now delete the log file again and check that it is not recreated: +#+begin_src shell + $ sudo rm -f /export2/guix-containers/genenetwork/uploader/var/log/gunicorn-g*.log + $ ls /export2/guix-containers/genenetwork/uploader/var/log/ +#+end_src +and we get +#+begin_example +debug maillog mcron.log.1.gz messages.1.gz nginx secure virtuoso.log +guix-daemon.log mcron.log messages mysqld.log nscd.log secure.1.gz wtmp +#+end_example + +Great success!!! 🎉🎉 + +Now, let us build the container with the pristine guix +#+begin_src shell + $ echo $PATH + /usr/local/bin:/usr/bin:/bin:/usr/local/games:/usr/games + $ export PATH="/usr/local/guix-profiles/guix-daemon/bin:${PATH}" + $ echo $PATH + /usr/local/guix-profiles/guix-daemon/bin:/usr/local/bin:/usr/bin:/bin:/usr/local/games:/usr/games + $ cd /home/fredm/gn-machines/ + $ ./uploader-deploy.sh /home/fredm/guix-forge /home/fredm/guix-past /home/fredm/guix-bioinformatics +#+end_src + +The upload step fails with +#+begin_example +===== Auxilliary module load paths ===== +-L /home/fredm/guix-forge/guix +-L /home/fredm/guix-past +-L /home/fredm/guix-bioinformatics +===== END: Auxilliary module load paths ===== +hint: Consider installing the `glibc-locales' package and defining +`GUIX_LOCPATH', along these lines: + + guix install glibc-locales + export GUIX_LOCPATH="$HOME/.guix-profile/lib/locale" + +See the "Application Setup" section in the manual, for more info. + +guix system: warning: Consider running 'guix pull' followed by +'guix system reconfigure' to get up-to-date packages and security updates. + +Backtrace: +In guix/store.scm: + 2065:12 19 (_ #<store-connection 256.100 7fae2bd9bd70>) + 1382:11 18 (map/accumulate-builds #<store-connection 256.100 7fae…> …) + 1300:8 17 (call-with-build-handler #<procedure 7fae2b7a4bd0 at g…> …) + 2180:25 16 (run-with-store #<store-connection 256.100 7fae2bd9bd70> …) +In guix/gexp.scm: + 914:13 15 (_ _) +In guix/store.scm: + 2008:8 14 (_ _) +In guix/gexp.scm: + 299:22 13 (_ _) +In guix/store.scm: + 2052:38 12 (_ #<store-connection 256.100 7fae2bd9bd70>) +In gnu/system.scm: + 1632:9 11 (_ _) +In guix/store.scm: + 2180:25 10 (run-with-store #<store-connection 256.100 7fae2bd9bd70> …) +In gnu/system.scm: + 1299:19 9 (_ _) + 836:11 8 (operating-system-services #<<operating-system> kernel:…>) +In gnu/system/linux-container.scm: + 174:27 7 (services _) +In ice-9/eval.scm: + 191:35 6 (_ #(#(#<module (#{ g709}#) 7fae2f601640>) #<<operati…>)) + 173:55 5 (_ #(#(#<module (#{ g709}#) 7fae2f601640>) #<<operati…>)) + 196:35 4 (_ #(#(#<module (#{ g709}#) 7fae2f601640>) #<<operati…>)) + 223:20 3 (proc #(#(#<module (#{ g709}#) 7fae2f601640>) #<<oper…>)) +In unknown file: + 2 (%resolve-variable (7 . genenetwork-service-type) #<mod…>) +In ice-9/boot-9.scm: + 1685:16 1 (raise-exception _ #:continuable? _) + 1685:16 0 (raise-exception _ #:continuable? _) + +ice-9/boot-9.scm:1685:16: In procedure raise-exception: +error: genenetwork-service-type: unbound variable +#+end_example + +😭😭😭 + +Reset path, and first build with non-pristine guix: +#+begin_src shell + $ export PATH="" + $ env PATH="/home/fredm/opt/guix/bin:${PATH}" ./uploader-deploy.sh +#+end_src +Success! + +Start the container +#+begin_src shell + sudo systemctl start genenetwork-uploader-container.service +#+end_src + +Check GN2 log for former weirdness +#+begin_src shell + sudo cat /export2/guix-containers/genenetwork/uploader/var/log/gunicorn-genenetwork2.log +#+end_src + +No more of the errors from the wrong profile! |