This runs logrotate as a background process rather than in the
foreground. Degradation was caused by a delay in container startup when
catalina.out gets rotated. If the file is big enough, the delay can be
significant to cause degradation.
Type of change
Please delete options that are not relevant.
[X] Bug fix (non-breaking change which fixes an issue)
How Has This Been Tested?
Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration
[X] Other - add any additional tests here
Testing is done by doing the following:
stop any existing metacat containers/processes in the case of Spin run this using ess-dive-config:
./deploy.sh dev stop data-dev-ess-dive-lbl/app-metacat
then check if catalina.out is big enough on the community file system located on Cori:
ls -lah $CFS/m2690/ess-dive/metacat/usr/local/tomcat/logs/catalina.out
if it is less than 100M then make it bigger by using truncate:
truncate -s 200M catalina.out
Then restart the container using the built image with the commit number p57. So modify the docker-compose.yml file in ess-dive-config. You'll have to build the image and push to the spin registry using the essdive-appstack repository.
./deploy.sh dev up -d app-metacat
Then check for the following:
private packages are visible
new data can be submitted
catalina.out gets logrotated.
Note: You will have to delete one of the old tarballs if they already exist since logrotate will check for the existence of that tarball. If it already exists then no logrotation will happen.
Background logrotate process
Description
Closes #22
This runs logrotate as a background process rather than in the foreground. Degradation was caused by a delay in container startup when catalina.out gets rotated. If the file is big enough, the delay can be significant to cause degradation.
Type of change
Please delete options that are not relevant.
How Has This Been Tested?
Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration
Testing is done by doing the following:
stop any existing metacat containers/processes in the case of Spin run this using
ess-dive-config:./deploy.sh dev stop data-dev-ess-dive-lbl/app-metacatthen check if catalina.out is big enough on the community file system located on Cori:
ls -lah $CFS/m2690/ess-dive/metacat/usr/local/tomcat/logs/catalina.outif it is less than 100M then make it bigger by using
truncate:truncate -s 200M catalina.outThen restart the container using the built image with the commit number
p57. So modify thedocker-compose.ymlfile iness-dive-config. You'll have to build the image and push to the spin registry using theessdive-appstackrepository../deploy.sh dev up -d app-metacatThen check for the following:
Note: You will have to delete one of the old tarballs if they already exist since logrotate will check for the existence of that tarball. If it already exists then no logrotation will happen.
Test Configuration