Open ElliotDavis opened 6 years ago
Hi Elliot,
Many thanks for founding this, we most of the time work on systemd, so the contribution of your script is very welcome !
Tx /svar
There is also a replication-manager-cli status command that could help to get the state of the server init script!
Hi Stephane,
I will forward it to you this weekend. I am in the middle of migrating away from Galera and to straight MariaDB master/slave. We are going to try and implement replication-manager-osc 2.0 but we have 2.0 stable in stage and monit keeps reporting the pid changes. Looking into it.
Also, graphs are not working in 2.0, well at least not for me even with it configured in the config.
Cheers --Elliot
On Wed, Mar 14, 2018 at 11:20 PM, Stephane VAROQUI <notifications@github.com
wrote:
Hi Elliot,
Many thanks for founding this, we most of the time work on systemd, so the contribution of your script is very welcome !
Tx /svar
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/218#issuecomment-373272821, or mute the thread https://github.com/notifications/unsubscribe-auth/AT-5B-lRiGCrsPCU5hFIurl9OurivTQ7ks5teggzgaJpZM4SrfVZ .
-- Cheers, Elliot Davis Tripping.com http://tripping.com/
Find your happy place on Tripping.com https://www.youtube.com/watch?v=i1n-z36VBzk
Hi, Elliot,
I’m able to get graph from GitHub 2.0 working using etc/config.toml.sample.tst.masterslave-graphite
graphite-metrics = true graphite-carbon-host = "127.0.0.1" graphite-carbon-port = 2003 graphite-embedded = true graphite-carbon-api-port = 10002 graphite-carbon-server-port = 10003 graphite-carbon-link-port = 7002 graphite-carbon-pickle-port = 2004 graphite-carbon-pprof-port = 7007
The graph.js is under vendor now
I strongly suggest to use grafana , those internal graph are very limited and need port 10002 to be open between the browser and repman server !
We do provide grafana template in share, they are graphing most important metrics!
Le 17 mars 2018 à 00:03, edavis notifications@github.com a écrit :
Hi Stephane,
I will forward it to you this weekend. I am in the middle of migrating away from Galera and to straight MariaDB master/slave. We are going to try and implement replication-manager-osc 2.0 but we have 2.0 stable in stage and monit keeps reporting the pid changes. Looking into it.
Also, graphs are not working in 2.0, well at least not for me even with it configured in the config.
Cheers --Elliot
On Wed, Mar 14, 2018 at 11:20 PM, Stephane VAROQUI <notifications@github.com
wrote:
Hi Elliot,
Many thanks for founding this, we most of the time work on systemd, so the contribution of your script is very welcome !
Tx /svar
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/218#issuecomment-373272821, or mute the thread https://github.com/notifications/unsubscribe-auth/AT-5B-lRiGCrsPCU5hFIurl9OurivTQ7ks5teggzgaJpZM4SrfVZ .
-- Cheers, Elliot Davis Tripping.com http://tripping.com/
Find your happy place on Tripping.com https://www.youtube.com/watch?v=i1n-z36VBzk — You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/218#issuecomment-373867774, or mute the thread https://github.com/notifications/unsubscribe-auth/AC1RIBh4MTCvosNBAoaDYeAYZBFEhjceks5tfETYgaJpZM4SrfVZ.
Stéphane Varoqui, VP of Products Phone: +33 695-926-401, skype: svaroqui https://signal18.io/ https://signal18.io/
Stephane, thanks for the fast response. I am using packages out of the repo for Ubuntu. I will do testing from the git hub on my ops machine and see if there is a difference. I wanted to say that so far this is being well received by our company. The hope is we can go to a traditional master/slave setup which is more performant than Galera and still give us the peace of mind through the night. My VP did want to ask about pro licensing and on-demand support was? I agree with Grafana, they use datadog here (Not sure if I like it). I playing getting MRM into influxdb, it should support it since it supports graphite protocol.
On Sat, Mar 17, 2018 at 1:32 AM, Stephane VAROQUI notifications@github.com wrote:
Hi, Elliot,
I’m able to get graph from GitHub 2.0 working using etc/config.toml.sample.tst.masterslave-graphite
graphite-metrics = true graphite-carbon-host = "127.0.0.1" graphite-carbon-port = 2003 graphite-embedded = true graphite-carbon-api-port = 10002 graphite-carbon-server-port = 10003 graphite-carbon-link-port = 7002 graphite-carbon-pickle-port = 2004 graphite-carbon-pprof-port = 7007
The graph.js is under vendor now
I strongly sugest to branch grafana , those internal graph are very limited
We do provide grafana template in shar, they are graphing most important metrics!
Le 17 mars 2018 à 00:03, edavis notifications@github.com a écrit :
Hi Stephane,
I will forward it to you this weekend. I am in the middle of migrating away from Galera and to straight MariaDB master/slave. We are going to try and implement replication-manager-osc 2.0 but we have 2.0 stable in stage and monit keeps reporting the pid changes. Looking into it.
Also, graphs are not working in 2.0, well at least not for me even with it configured in the config.
Cheers --Elliot
On Wed, Mar 14, 2018 at 11:20 PM, Stephane VAROQUI < notifications@github.com
wrote:
Hi Elliot,
Many thanks for founding this, we most of the time work on systemd, so the contribution of your script is very welcome !
Tx /svar
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/ 218#issuecomment-373272821, or mute the thread https://github.com/notifications/unsubscribe-auth/AT-5B- lRiGCrsPCU5hFIurl9OurivTQ7ks5teggzgaJpZM4SrfVZ .
-- Cheers, Elliot Davis Tripping.com http://tripping.com/
Find your happy place on Tripping.com https://www.youtube.com/watch?v=i1n-z36VBzk — You are receiving this because you commented. Reply to this email directly, view it on GitHub < https://github.com/signal18/replication-manager/issues/ 218#issuecomment-373867774>, or mute the thread https://github.com/ notifications/unsubscribe-auth/AC1RIBh4MTCvosNBAoaDYeAYZBFEhj ceks5tfETYgaJpZM4SrfVZ.
Stéphane Varoqui, VP of Products Phone: +33 695-926-401, skype: svaroqui https://signal18.io/ https://signal18.io/
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/218#issuecomment-373903803, or mute the thread https://github.com/notifications/unsubscribe-auth/AT-5BySBTdUe5CQA4FGMFpFik0lQ_BcYks5tfModgaJpZM4SrfVZ .
-- Cheers, Elliot Davis Tripping.com http://tripping.com/
Find your happy place on Tripping.com https://www.youtube.com/watch?v=i1n-z36VBzk
Re,
That’s probably better to set our depot
What release of ubuntu are you on ?
Latests release in constant integration can be found here http://ci.signal18.io/mrm/builds/ http://ci.signal18.io/mrm/builds/
Tx /svar
Le 17 mars 2018 à 10:06, edavis notifications@github.com a écrit :
Stephane, thanks for the fast response. I am using packages out of the repo for Ubuntu. I will do testing from the git hub on my ops machine and see if there is a difference. I wanted to say that so far this is being well received by our company. The hope is we can go to a traditional master/slave setup which is more performant than Galera and still give us the peace of mind through the night. My VP did want to ask about pro licensing and on-demand support was? I agree with Grafana, they use datadog here (Not sure if I like it). I playing getting MRM into influxdb, it should support it since it supports graphite protocol.
On Sat, Mar 17, 2018 at 1:32 AM, Stephane VAROQUI notifications@github.com wrote:
Hi, Elliot,
I’m able to get graph from GitHub 2.0 working using etc/config.toml.sample.tst.masterslave-graphite
graphite-metrics = true graphite-carbon-host = "127.0.0.1" graphite-carbon-port = 2003 graphite-embedded = true graphite-carbon-api-port = 10002 graphite-carbon-server-port = 10003 graphite-carbon-link-port = 7002 graphite-carbon-pickle-port = 2004 graphite-carbon-pprof-port = 7007
The graph.js is under vendor now
I strongly sugest to branch grafana , those internal graph are very limited
We do provide grafana template in shar, they are graphing most important metrics!
Le 17 mars 2018 à 00:03, edavis notifications@github.com a écrit :
Hi Stephane,
I will forward it to you this weekend. I am in the middle of migrating away from Galera and to straight MariaDB master/slave. We are going to try and implement replication-manager-osc 2.0 but we have 2.0 stable in stage and monit keeps reporting the pid changes. Looking into it.
Also, graphs are not working in 2.0, well at least not for me even with it configured in the config.
Cheers --Elliot
On Wed, Mar 14, 2018 at 11:20 PM, Stephane VAROQUI < notifications@github.com
wrote:
Hi Elliot,
Many thanks for founding this, we most of the time work on systemd, so the contribution of your script is very welcome !
Tx /svar
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/ 218#issuecomment-373272821, or mute the thread https://github.com/notifications/unsubscribe-auth/AT-5B- lRiGCrsPCU5hFIurl9OurivTQ7ks5teggzgaJpZM4SrfVZ .
-- Cheers, Elliot Davis Tripping.com http://tripping.com/
Find your happy place on Tripping.com https://www.youtube.com/watch?v=i1n-z36VBzk — You are receiving this because you commented. Reply to this email directly, view it on GitHub < https://github.com/signal18/replication-manager/issues/ 218#issuecomment-373867774>, or mute the thread https://github.com/ notifications/unsubscribe-auth/AC1RIBh4MTCvosNBAoaDYeAYZBFEhj ceks5tfETYgaJpZM4SrfVZ.
Stéphane Varoqui, VP of Products Phone: +33 695-926-401, skype: svaroqui https://signal18.io/ https://signal18.io/
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/218#issuecomment-373903803, or mute the thread https://github.com/notifications/unsubscribe-auth/AT-5BySBTdUe5CQA4FGMFpFik0lQ_BcYks5tfModgaJpZM4SrfVZ .
-- Cheers, Elliot Davis Tripping.com http://tripping.com/
Find your happy place on Tripping.com https://www.youtube.com/watch?v=i1n-z36VBzk — You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/signal18/replication-manager/issues/218#issuecomment-373905635, or mute the thread https://github.com/notifications/unsubscribe-auth/AC1RILcWXK0zKi3PD6U1LJqKhFMWyqEyks5tfNH3gaJpZM4SrfVZ.
Stéphane Varoqui, VP of Products Phone: +33 695-926-401, skype: svaroqui https://signal18.io/ https://signal18.io/
Hello,
Can you request by email price list at stephane@signal18.io
Cheers /Svar
The init.d script packaged in replication-manager-osc 2.0/2.1 leaves orphaned process on the server. The issue comes from the pid file being updated with a null value and the server is unable to shut down the process. It will report it is stopped, however ps aux | grep replication still shows an active running process. I have rewritten the init script and tested it in our environment which is under chef control. Please let me know if you would like a copy of the init.d script I modified.