Re: EPEL - rb-site upgrade failure upon reboot prevents httpd startup
Hi Stephen,
Your fix has resolved the issue on my system. Thanks for the rapid
response.
Cheers,
Paul
On Wednesday, October 29, 2014 11:39:53 PM UTC, Stephen Gallagher wrote:
On Wed, 2014-10-29 at 16:27 -0400, Stephen Gallagher wrote:
On Wed, 2014-10-29 at 11:04 -0700, Paul Fee wrote:
Hi all (and Stephen Gallagher),
I'm using CentOS7 + EPEL with ReviewBoard-2.0.11 (fresh install +
today's updates). My site is listed in /etc/reviewboard/sites. When
httpd starts up, it checks if any sites need upgraded. I'm using a
mariadb database.
When I reboot the box, httpd fails to startup.
$ sudo systemctl status httpd
httpd.service - The Apache HTTP Server
Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
Drop-In: /usr/lib/systemd/system/httpd.service.d
└─reviewboard-sites.conf
Active: failed (Result: exit-code) since Wed 2014-10-29 17:42:28
GMT; 16s ago
Process: 1184 ExecStartPre=/usr/bin/rb-site upgrade --all-sites
(code=exited, status=1/FAILURE)
Oct 29 17:42:28 centos7 rb-site[1184]: File
/usr/lib/python2.7/site-packages/django/db/backends/mysql/base.py,
line 435, in get_new_connection
Oct 29 17:42:28 centos7 rb-site[1184]: conn =
Database.connect(**conn_params)
Oct 29 17:42:28 centos7 rb-site[1184]: File
/usr/lib64/python2.7/site-packages/MySQLdb/__init__.py, line 81, in
Connect
Oct 29 17:42:28 centos7 rb-site[1184]: return Connection(*args,
**kwargs)
Oct 29 17:42:28 centos7 rb-site[1184]: File
/usr/lib64/python2.7/site-packages/MySQLdb/connections.py, line 187,
in __init__
Oct 29 17:42:28 centos7 rb-site[1184]: super(Connection,
self).__init__(*args, **kwargs2)
Oct 29 17:42:28 centos7 rb-site[1184]:
django.db.utils.OperationalError: (2002, Can't connect to local MySQL
server through socket '...k' (2))
Oct 29 17:42:28 centos7 systemd[1]: httpd.service: control process
exited, code=exited status=1
Oct 29 17:42:28 centos7 systemd[1]: Failed to start The Apache HTTP
Server.
Oct 29 17:42:28 centos7 systemd[1]: Unit httpd.service entered failed
state.
Hint: Some lines were ellipsized, use -l to show in full.
==
I can manually start httpd as mariadb is now running by the time I've
logged in.
If I remove the /etc/reviewboard/sites file, the issue goes away. If
I restore the file, the issue comes back.
It looks like rb-site upgrade is being run before the database is
available. Should systemd be using socket activation to start mariadb
once a connection attempt is made? I believe
the /etc/reviewboard/sites facility is unique to Fedora/EPEL, so this
is could be a distro issue rather than ReviewBoard.
MariaDB does not currently support socket-activation. What you need to
do as a workaround right now is to add the following lines
to /usr/lib/systemd/system/httpd.service.d/reviewboard-sites.conf:
[Unit]
After=postgresql.service mariadb.service mysql.service
I'll roll this into the Fedora and EPEL packages. I hadn't thought of it
previously. (Note: it's safe to list all of them on the After= line as
this only controls ordering; it doesn't cause any of them to be started
that would not already have been).
Now I just have to hope that Apache never grows a dependency on a
database... :)
Fedora 21:
https://admin.fedoraproject.org/updates/FEDORA-2014-13850/ReviewBoard-2.0.11-2.fc21
EPEL 7:
https://admin.fedoraproject.org/updates/ReviewBoard-2.0.11-2.el7
--
Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
---
Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
---
Happy user? Let us know at http://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups
reviewboard group.
To unsubscribe from this group and stop receiving emails from it, send an email
to reviewboard+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
EPEL - rb-site upgrade failure upon reboot prevents httpd startup
Hi all (and Stephen Gallagher), I'm using CentOS7 + EPEL with ReviewBoard-2.0.11 (fresh install + today's updates). My site is listed in /etc/reviewboard/sites. When httpd starts up, it checks if any sites need upgraded. I'm using a mariadb database. When I reboot the box, httpd fails to startup. $ sudo systemctl status httpd httpd.service - The Apache HTTP Server Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled) Drop-In: /usr/lib/systemd/system/httpd.service.d └─reviewboard-sites.conf Active: failed (Result: exit-code) since Wed 2014-10-29 17:42:28 GMT; 16s ago Process: 1184 ExecStartPre=/usr/bin/rb-site upgrade --all-sites (code=exited, status=1/FAILURE) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib/python2.7/site-packages/django/db/backends/mysql/base.py, line 435, in get_new_connection Oct 29 17:42:28 centos7 rb-site[1184]: conn = Database.connect(**conn_params) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib64/python2.7/site-packages/MySQLdb/__init__.py, line 81, in Connect Oct 29 17:42:28 centos7 rb-site[1184]: return Connection(*args, **kwargs) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib64/python2.7/site-packages/MySQLdb/connections.py, line 187, in __init__ Oct 29 17:42:28 centos7 rb-site[1184]: super(Connection, self).__init__(*args, **kwargs2) Oct 29 17:42:28 centos7 rb-site[1184]: django.db.utils.OperationalError: (2002, Can't connect to local MySQL server through socket '...k' (2)) Oct 29 17:42:28 centos7 systemd[1]: httpd.service: control process exited, code=exited status=1 Oct 29 17:42:28 centos7 systemd[1]: Failed to start The Apache HTTP Server. Oct 29 17:42:28 centos7 systemd[1]: Unit httpd.service entered failed state. Hint: Some lines were ellipsized, use -l to show in full. == I can manually start httpd as mariadb is now running by the time I've logged in. If I remove the /etc/reviewboard/sites file, the issue goes away. If I restore the file, the issue comes back. It looks like rb-site upgrade is being run before the database is available. Should systemd be using socket activation to start mariadb once a connection attempt is made? I believe the /etc/reviewboard/sites facility is unique to Fedora/EPEL, so this is could be a distro issue rather than ReviewBoard. Thanks, Paul -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups reviewboard group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: EPEL - rb-site upgrade failure upon reboot prevents httpd startup
On Wed, 2014-10-29 at 11:04 -0700, Paul Fee wrote: Hi all (and Stephen Gallagher), I'm using CentOS7 + EPEL with ReviewBoard-2.0.11 (fresh install + today's updates). My site is listed in /etc/reviewboard/sites. When httpd starts up, it checks if any sites need upgraded. I'm using a mariadb database. When I reboot the box, httpd fails to startup. $ sudo systemctl status httpd httpd.service - The Apache HTTP Server Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled) Drop-In: /usr/lib/systemd/system/httpd.service.d └─reviewboard-sites.conf Active: failed (Result: exit-code) since Wed 2014-10-29 17:42:28 GMT; 16s ago Process: 1184 ExecStartPre=/usr/bin/rb-site upgrade --all-sites (code=exited, status=1/FAILURE) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib/python2.7/site-packages/django/db/backends/mysql/base.py, line 435, in get_new_connection Oct 29 17:42:28 centos7 rb-site[1184]: conn = Database.connect(**conn_params) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib64/python2.7/site-packages/MySQLdb/__init__.py, line 81, in Connect Oct 29 17:42:28 centos7 rb-site[1184]: return Connection(*args, **kwargs) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib64/python2.7/site-packages/MySQLdb/connections.py, line 187, in __init__ Oct 29 17:42:28 centos7 rb-site[1184]: super(Connection, self).__init__(*args, **kwargs2) Oct 29 17:42:28 centos7 rb-site[1184]: django.db.utils.OperationalError: (2002, Can't connect to local MySQL server through socket '...k' (2)) Oct 29 17:42:28 centos7 systemd[1]: httpd.service: control process exited, code=exited status=1 Oct 29 17:42:28 centos7 systemd[1]: Failed to start The Apache HTTP Server. Oct 29 17:42:28 centos7 systemd[1]: Unit httpd.service entered failed state. Hint: Some lines were ellipsized, use -l to show in full. == I can manually start httpd as mariadb is now running by the time I've logged in. If I remove the /etc/reviewboard/sites file, the issue goes away. If I restore the file, the issue comes back. It looks like rb-site upgrade is being run before the database is available. Should systemd be using socket activation to start mariadb once a connection attempt is made? I believe the /etc/reviewboard/sites facility is unique to Fedora/EPEL, so this is could be a distro issue rather than ReviewBoard. MariaDB does not currently support socket-activation. What you need to do as a workaround right now is to add the following lines to /usr/lib/systemd/system/httpd.service.d/reviewboard-sites.conf: [Unit] After=postgresql.service mariadb.service mysql.service I'll roll this into the Fedora and EPEL packages. I hadn't thought of it previously. (Note: it's safe to list all of them on the After= line as this only controls ordering; it doesn't cause any of them to be started that would not already have been). Now I just have to hope that Apache never grows a dependency on a database... :) -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups reviewboard group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: EPEL - rb-site upgrade failure upon reboot prevents httpd startup
On Wed, 2014-10-29 at 16:27 -0400, Stephen Gallagher wrote: On Wed, 2014-10-29 at 11:04 -0700, Paul Fee wrote: Hi all (and Stephen Gallagher), I'm using CentOS7 + EPEL with ReviewBoard-2.0.11 (fresh install + today's updates). My site is listed in /etc/reviewboard/sites. When httpd starts up, it checks if any sites need upgraded. I'm using a mariadb database. When I reboot the box, httpd fails to startup. $ sudo systemctl status httpd httpd.service - The Apache HTTP Server Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled) Drop-In: /usr/lib/systemd/system/httpd.service.d └─reviewboard-sites.conf Active: failed (Result: exit-code) since Wed 2014-10-29 17:42:28 GMT; 16s ago Process: 1184 ExecStartPre=/usr/bin/rb-site upgrade --all-sites (code=exited, status=1/FAILURE) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib/python2.7/site-packages/django/db/backends/mysql/base.py, line 435, in get_new_connection Oct 29 17:42:28 centos7 rb-site[1184]: conn = Database.connect(**conn_params) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib64/python2.7/site-packages/MySQLdb/__init__.py, line 81, in Connect Oct 29 17:42:28 centos7 rb-site[1184]: return Connection(*args, **kwargs) Oct 29 17:42:28 centos7 rb-site[1184]: File /usr/lib64/python2.7/site-packages/MySQLdb/connections.py, line 187, in __init__ Oct 29 17:42:28 centos7 rb-site[1184]: super(Connection, self).__init__(*args, **kwargs2) Oct 29 17:42:28 centos7 rb-site[1184]: django.db.utils.OperationalError: (2002, Can't connect to local MySQL server through socket '...k' (2)) Oct 29 17:42:28 centos7 systemd[1]: httpd.service: control process exited, code=exited status=1 Oct 29 17:42:28 centos7 systemd[1]: Failed to start The Apache HTTP Server. Oct 29 17:42:28 centos7 systemd[1]: Unit httpd.service entered failed state. Hint: Some lines were ellipsized, use -l to show in full. == I can manually start httpd as mariadb is now running by the time I've logged in. If I remove the /etc/reviewboard/sites file, the issue goes away. If I restore the file, the issue comes back. It looks like rb-site upgrade is being run before the database is available. Should systemd be using socket activation to start mariadb once a connection attempt is made? I believe the /etc/reviewboard/sites facility is unique to Fedora/EPEL, so this is could be a distro issue rather than ReviewBoard. MariaDB does not currently support socket-activation. What you need to do as a workaround right now is to add the following lines to /usr/lib/systemd/system/httpd.service.d/reviewboard-sites.conf: [Unit] After=postgresql.service mariadb.service mysql.service I'll roll this into the Fedora and EPEL packages. I hadn't thought of it previously. (Note: it's safe to list all of them on the After= line as this only controls ordering; it doesn't cause any of them to be started that would not already have been). Now I just have to hope that Apache never grows a dependency on a database... :) Fedora 21: https://admin.fedoraproject.org/updates/FEDORA-2014-13850/ReviewBoard-2.0.11-2.fc21 EPEL 7: https://admin.fedoraproject.org/updates/ReviewBoard-2.0.11-2.el7 -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups reviewboard group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
