PR for https://github.com/389ds/389-ds-base/issues/2790 committed again to 389-ds. So, with recently released 1.4.3.28 (the only release having this change for now) IPA Azure CI is expectedly broken:
Oct 28 14:31:11 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:11.881637118 +0000] - INFO - main - 389-Directory/1.4.3.28 B2021.301.0900 starting up Oct 28 14:31:11 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:11.882426833 +0000] - INFO - main - Setting the maximum file descriptor limit to: 1048576 Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.890639645 +0000] - INFO - PBKDF2_SHA256 - Based on CPU performance, chose 2048 rounds Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.897645571 +0000] - INFO - bdb_config_upgrade_dse_info - create config entry from old config Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.906162424 +0000] - NOTICE - ldbm_back_start - found 3276800k physical memory Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.907040140 +0000] - NOTICE - ldbm_back_start - found 2427124k available Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.907800253 +0000] - NOTICE - ldbm_back_start - cache autosizing: db cache: 204800k Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.908430665 +0000] - NOTICE - ldbm_back_start - total cache size: 167772160 B; Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.909666187 +0000] - ERR - no_diskspace - No enough space left on device (/dev/shm/slapd-IPA-TEST) (67100672 bytes); at least 184549376 bytes space is needed for db region files Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.910216597 +0000] - ERR - ldbm_back_start - Failed to init database, err=28 No space left on device Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.910699005 +0000] - ERR - plugin_dependency_startall - Failed to start database plugin ldbm database Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.916052402 +0000] - CRIT - dblayer_setup - dblayer_init failed Oct 28 14:31:12 master1.ipa.test ns-slapd[4593]: [28/Oct/2021:14:31:12.916911117 +0000] - ERR - ldbm_back_start - Failed to setup dblayer
IPA upstream CI is not affected at the moment, but eventually, this will happen unless it is reverted back.
For example, 389-ds's CI set 4gb for /dev/shm for Docker container via --shm-size="":
--shm-size=""
Size of /dev/shm. The format is . number must be greater than 0. Unit is optional and can be b (bytes), k (kilobytes), m (megabytes), or g (gigabytes). If you omit the unit, the system uses bytes. If you omit the size entirely, the system uses 64m.
DS team will modify the installer to not set the db home directory when inside a container.
Marking as a tracker since the changes will be done in 389-.ds
Metadata Update from @rcritten: - Issue tagged with: tracker
The 389-ds issue was resolved on Jan 31, 2022. Marking as closed.
Metadata Update from @rcritten: - Issue close_status updated to: fixed - Issue status updated to: Closed (was: Open)