)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":28223,"name":"Cedric Jeanneret","display_name":"cjeanner (Tengu)","email":"cjeanner@redhat.com","username":"cjeanner"},"change_message_id":"a30dd14d40f1aff96fba3c824f9daeced00a148f","unresolved":true,"context_lines":[{"line_number":15,"context_line":"systemd-firstboot.service actually runs, and other units can use"},{"line_number":16,"context_line":"first-boot-complete.target as a dependency to trigger on first boot."},{"line_number":17,"context_line":""},{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"68cdfc13_2d9b22ab","line":18,"range":{"start_line":18,"start_character":6,"end_line":18,"end_character":30},"updated":"2022-04-11 05:35:56.000000000","message":"afaik, only for older releases - el9 as well as el8 don\u0027t seem to have it anymore","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"ec70373d3cd9ec73351b10f7d1858528bcf7c767","unresolved":true,"context_lines":[{"line_number":15,"context_line":"systemd-firstboot.service actually runs, and other units can use"},{"line_number":16,"context_line":"first-boot-complete.target as a dependency to trigger on first boot."},{"line_number":17,"context_line":""},{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"ce1d8522_5ee64ef1","line":18,"range":{"start_line":18,"start_character":6,"end_line":18,"end_character":30},"in_reply_to":"68cdfc13_2d9b22ab","updated":"2022-04-11 22:24:05.000000000","message":"the -e check will handle /var/lib/dbus/machine-id not being there.","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":28223,"name":"Cedric Jeanneret","display_name":"cjeanner (Tengu)","email":"cjeanner@redhat.com","username":"cjeanner"},"change_message_id":"9b0d00c39afe8825d79e2f08383e87ebeeaf45d9","unresolved":false,"context_lines":[{"line_number":15,"context_line":"systemd-firstboot.service actually runs, and other units can use"},{"line_number":16,"context_line":"first-boot-complete.target as a dependency to trigger on first boot."},{"line_number":17,"context_line":""},{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"2f84c118_c46341f7","line":18,"range":{"start_line":18,"start_character":6,"end_line":18,"end_character":30},"in_reply_to":"ce1d8522_5ee64ef1","updated":"2022-04-12 04:45:43.000000000","message":"Ack","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":28223,"name":"Cedric Jeanneret","display_name":"cjeanner (Tengu)","email":"cjeanner@redhat.com","username":"cjeanner"},"change_message_id":"a30dd14d40f1aff96fba3c824f9daeced00a148f","unresolved":true,"context_lines":[{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Change-Id: I77c35e51a3da2e8a6b5a2c80d033a159b303c9af"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"75eb752f_58a3befe","line":21,"updated":"2022-04-11 05:35:56.000000000","message":"Well, it\u0027s not that clear:\n\"If by any of the above rules, a first boot is detected, units with ConditionFirstBoot\u003dyes will be run.\"\n\nSooo... empty file is still triggering ConditionFirstBoot\u003dyes things.\n\nBut it\u0027s probably better to be on the safest side, that machine-id thing is one of those over-engineered things anyway.\nNote the \"/etc/machine-info\" that exists now, showing yet another machine-id - though it\u0027s apparently not always present, and apparently has no actual impact on the real machine-id content.","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"a47cc1b2b2814044f9d49a8044526c19672c8f29","unresolved":true,"context_lines":[{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Change-Id: I77c35e51a3da2e8a6b5a2c80d033a159b303c9af"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"eab7d8dc_8a0a7765","line":21,"in_reply_to":"0d00f7f1_ca434ab1","updated":"2022-04-12 20:28:19.000000000","message":"The machine-id docs are correct I think, they say this which matches with my observed behaviour:\n\n  If /etc/machine-id exists and is empty, a boot is not considered the first boot. systemd will still bind-mount a file containing the actual machine-id over it and later try to commit it to disk (if /etc/ is writable).","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"ec70373d3cd9ec73351b10f7d1858528bcf7c767","unresolved":true,"context_lines":[{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Change-Id: I77c35e51a3da2e8a6b5a2c80d033a159b303c9af"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"dd826d52_ae6fcf25","line":21,"in_reply_to":"75eb752f_58a3befe","updated":"2022-04-11 22:24:05.000000000","message":"\u003e Sooo... empty file is still triggering ConditionFirstBoot\u003dyes things.\n\nNot according to my testing on centos9 :(\n\nI can see in the journal [systemd] generating the machine-id, but units with ConditionFirstBoot\u003dyes do not run. The context for this change is https://review.opendev.org/c/openstack/tripleo-image-elements/+/837430","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":28223,"name":"Cedric Jeanneret","display_name":"cjeanner (Tengu)","email":"cjeanner@redhat.com","username":"cjeanner"},"change_message_id":"9b0d00c39afe8825d79e2f08383e87ebeeaf45d9","unresolved":true,"context_lines":[{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Change-Id: I77c35e51a3da2e8a6b5a2c80d033a159b303c9af"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"0d00f7f1_ca434ab1","line":21,"in_reply_to":"dd826d52_ae6fcf25","updated":"2022-04-12 04:45:43.000000000","message":"if the doc isn\u0027t correct (or, well, the implementation)... That\u0027s a bug that should be reported then (against systemd?) imho. The conditions listed in the doc are precise and really tell any of them would trigger an actual FirstBoot :/","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":28223,"name":"Cedric Jeanneret","display_name":"cjeanner (Tengu)","email":"cjeanner@redhat.com","username":"cjeanner"},"change_message_id":"400649fb6521aa62e343e64c5df5636142340454","unresolved":false,"context_lines":[{"line_number":18,"context_line":"Since /var/lib/dbus/machine-id is sometimes a symlink to"},{"line_number":19,"context_line":"/etc/machine-id, it is truncated before writing to /etc/machine-id."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"[1] https://www.freedesktop.org/software/systemd/man/machine-id.html"},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Change-Id: I77c35e51a3da2e8a6b5a2c80d033a159b303c9af"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"a3a0a8f2_e4007437","line":21,"in_reply_to":"eab7d8dc_8a0a7765","updated":"2022-04-13 04:29:19.000000000","message":"/me lets coffee kick in..... *ding*!\n\nyou\u0027re right. But it sounds so wrong :(. So basically, the \"workaround\" for the virt-sysprep and virt-customize aren\u0027t correct (add --truncate /etc/machine-id).... How NOT convenient. Guess I\u0027ll have to rework a bit my things in the lab.","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"}],"/PATCHSET_LEVEL":[{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"7da699342a3e4d32311d551d31b93c32bb2959aa","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"644c7608_f165d2ac","updated":"2022-04-19 07:06:55.000000000","message":"I kind of wondered why I\u0027d not heard of this because I remember reading that page, but time flies 😊\n\nAFAICS this came in with\n\nhttps://github.com/systemd/systemd/commit/3023f2fead7df9f219acf3c8595d974e31ad23cc\n\nIf I\u0027m reading that right, that means systemd 247 is the first release that supports this?\n\ncentos 8 seems to be at 239; focal 245 ... so we have a lot of distros that won\u0027t know what is going on?\n\nI\u0027m concerned that these are all thinking their machine-id is \"uninitalized\" and that doesn\u0027t matter in testing here where everything is ephemeral, but may cause issues when we bring up more hosts? \n\nI\u0027m not 100% sure it\u0027s *not* backwards compatible, but reading around I couldn\u0027t see anything say it explicitly was.  Only -1 as I think this is worth double-checking and noting one way or the other ... if it\u0027s not, probably a \"systemctl --version\" check would work?","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"1a4641dd7e11b02a43ff1486936b8414b85b44ba","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"b2825f1e_59c82a57","in_reply_to":"331b9e9c_f330b46e","updated":"2022-04-20 05:36:04.000000000","message":"I chased it down a little further ...\n\nThe acutal code seems to be https://github.com/systemd/systemd/commit/583cef3b7347c4e6ca269d38efef6d2e4314aba6 ... not the other one i said\n\nI did double check at https://git.centos.org/rpms/systemd/blob/c7/f/SOURCES none of this got backported\n\nThe bit that has been there for many years is https://github.com/systemd/systemd/blob/da13d2ca0731b413841663052f2cc6832a855334/src/shared/machine-id-setup.c#L130 which generates a fresh machine-id if it reads something that isn\u0027t a 128-bit uuid.  So this must be what it has fallen into?\n\nI\u0027m just a bit worried that the existence of a file might have other issues.  AFAICS from the logs, the problems we\u0027ve had before have not so much been the existence of the file, but that they had a uuid in them already.  so maybe \"invalid value\" \u003d\u003d \"no file\"?  if we\u0027re convinced it is, i guess we can go with this.  we should probably make a note inline about this though.","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"e68faf3700a679607ff81162a472eee5e763ead0","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"82de7998_d62eae86","in_reply_to":"4be1dd03_d6eeb90e","updated":"2022-04-27 22:55:14.000000000","message":"Done","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"d06d44c7df1741903163843f2e5d830fced6e325","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"331b9e9c_f330b46e","in_reply_to":"644c7608_f165d2ac","updated":"2022-04-20 02:02:14.000000000","message":"I\u0027ve just checked the behaviour on centos7/systemd-219, it looks like the \"uninitalized\" behaviour itself has existed for some time and the above commit makes use of that:\n\n  $ echo uninitialized \u003e /etc/machine-id\n  $ reboot\n  ...\n  $ cat /etc/machine-id \n  76686998bcc84f079f661c1d29f03518\n  $ systemctl --version\n  systemd 219\n  +PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT \n  +GNUTLS +ACL +XZ +LZ4 -SECCOMP +BLKID +ELFUTILS +KMOD +IDN\n  $ cat /etc/redhat-release \n  CentOS Linux release 7.8.2003 (Core","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"46bb0e49d8f226f62c0521aa0bbb7a8e9b024c0a","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"4be1dd03_d6eeb90e","in_reply_to":"b2825f1e_59c82a57","updated":"2022-04-27 22:49:37.000000000","message":"I\u0027ll follow up with a note about why this works on older vs newer systemd.\n\nRemember though, the current behaviour is to truncate /etc/machine-id, rather than no file. So in relation to the documented first boot semantics, we were triggering 3., now we\u0027re triggering 2.\n\n[1] https://www.freedesktop.org/software/systemd/man/machine-id.html#First%20Boot%20Semantics","commit_id":"2c485f4f495dd31ca3432e61c19929c1eac361ab"},{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"0319981a8a4c9b7cc73da0c8cbcf43be8ae77a7e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"bcc94483_23cf0fd1","updated":"2022-04-22 04:20:53.000000000","message":"I\u0027m happy with the note and what we\u0027ve looked into.  If we do have an issue, we\u0027ve got a good head-start on it","commit_id":"147641fc3e11602cf9eaf723a2b38f60f394ac0b"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"46bb0e49d8f226f62c0521aa0bbb7a8e9b024c0a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"3a22c2d4_e8a35115","updated":"2022-04-27 22:49:37.000000000","message":"recheck","commit_id":"147641fc3e11602cf9eaf723a2b38f60f394ac0b"},{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"6f4b7a4420afba6f28ed87b2e863860e1f57a09c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"bbe4fb1c_b9582363","updated":"2022-04-26 04:03:53.000000000","message":"recheck\n\nnot sure why the suse image test failed.  made me note that we\u0027re checking the checksum after we use the downloaded file; https://review.opendev.org/c/openstack/diskimage-builder/+/839294","commit_id":"147641fc3e11602cf9eaf723a2b38f60f394ac0b"}]}