Right. My thinking was that if TS is the one responsible for aggregating that
code, it can generate a SHA of all those index definitions, no?
class User < AR
define_index do
indexes :username
….
end
end
The define_index interpreted and built with TS right? So I would assume TS can
loop through all its defined indexes and hash it somehow…
I'm probably way off so sorry if I'm just talking nonsense :-)
Just an idea…
On Monday, June 25, 2012 at 2:04 PM, Pat Allan wrote:
> But that won't work, because what you need to compare is the code in the
> define_index blocks. If those index definitions change (or you add or remove
> an index on a model), then you need to run ts:rebuild. But if there's been no
> change to the definitions, then you shouldn't need to restart or rebuild.
>
> --
> Pat
>
> On 25/06/2012, at 11:01 PM, Ngan Pham wrote:
>
> > I think we're misunderstanding each other…
> >
> > Currently when we deploy our application, we always do a TS restart. If we
> > were to remove the TS restart step, a developer would have to know that a
> > TS index was changed in the code and restart TS within the deployment
> > procedure manually. I would be great if the deployment process can figure
> > out if TS needs to be restarted or not basic on whether or not indexes were
> > changed since the last deployment. I was thinking that a way of doing this
> > would be to SHA the all the indexes that TS has and compare them for each
> > deployment. If the SHA has changed, that means the indexes have changed and
> > a TS restart is required.
> >
> > On Monday, June 25, 2012 at 1:33 PM, Pat Allan wrote:
> >
> > > How would you do that from a database migrations perspective?
> > >
> > > --
> > > Pat
> > >
> > > On 25/06/2012, at 10:29 PM, Ngan Pham wrote:
> > >
> > > > As you suggested that it's not necessary to restart TS every deploy
> > > > except when making index changes. I just wanted to automate the process
> > > > so a developer would have to remember to restart TS or not. Would be
> > > > nice for the deploy process to see that the indexes have changed and
> > > > restart TS automatically.
> > > > On Monday, June 25, 2012 at 1:24 PM, Pat Allan wrote:
> > > >
> > > > > Well, you could SHA or MD5 the files to easily note whether they've
> > > > > changed - but that won't tell you whether it's just the data that's
> > > > > changed, or the structure of them - so it's not useful as an
> > > > > indicator of whether you need to run ts:rebuild as part of a deploy,
> > > > > if that's what you were thinking?
> > > > >
> > > > > --
> > > > > Pat
> > > > >
> > > > > On 25/06/2012, at 6:46 PM, Ngan wrote:
> > > > >
> > > > > > Is there a reliable, programatic way of determining if your indexes
> > > > > > have changed? E.g. a SHA of the indexes?
> > > > > >
> > > > > > On Friday, June 8, 2012 8:52:32 AM UTC-7, Pat Allan wrote:
> > > > > > Yeah, if you keep things to the shared paths, then you should be
> > > > > > fine. Those docs certainly need some work (most of my deployments
> > > > > > are to Heroku these days, so I'm a bit out of practice with
> > > > > > Capistrano).
> > > > > > --
> > > > > > Pat
> > > > > >
> > > > > > On 08/06/2012, at 5:50 PM, Michael Guterl wrote:
> > > > > >
> > > > > > > Nope. I was just following the instructions detailed here:
> > > > > > >
> > > > > > > http://freelancing-god.github.com/ts/en/deployment.html
> > > > > > >
> > > > > > > I was thinking I only need to symlink, but I probably don't even
> > > > > > > need to do that seeing that the sphinx.yml points to the shared
> > > > > > > path. Do you agree?
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Michael Guterl
> > > > > > >
> > > > > > > On Thursday, June 7, 2012 3:55:23 PM UTC-4, Pat Allan wrote:
> > > > > > > Hi Michael
> > > > > > > Are you using ts:rebuild? You probably don't need to do that -
> > > > > > > indeed, treat the ts:rebuild task like db:migrate: it's only
> > > > > > > necessary when you add/remove indices or make changes to their
> > > > > > > structure. There's no need to restart Sphinx on every deploy.
> > > > > > >
> > > > > > > Cheers
> > > > > > >
> > > > > > > --
> > > > > > > Pat
> > > > > > >
> > > > > > > On 07/06/2012, at 9:39 PM, Michael Guterl wrote:
> > > > > > >
> > > > > > > > Currently when I deploy, searchd is restarted which causes my
> > > > > > > > application to raise errors if someone is trying to search or
> > > > > > > > if a resque process is using searchd. I'm using the capistrano
> > > > > > > > tasks included in thinking-sphinx and I'm wondering if there's
> > > > > > > > any way to achieve a rolling (or zero-downtime) deploy when
> > > > > > > > using these tasks?
> > > > > > > >
> > > > > > > > Am I doing something wrong?
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Michael Guterl
> > > > > > > >
> > > > > > > > --
> > > > > > > > You received this message because you are subscribed to the
> > > > > > > > Google Groups "Thinking Sphinx" group.
> > > > > > > > To view this discussion on the web visit
> > > > > > > > https://groups.google.com/d/msg/thinking-sphinx/-/yXAAtZ0hRkQJ.
> > > > > > > > To post to this group, send email to
> > > > > > > > [email protected]
> > > > > > > > (mailto:[email protected]).
> > > > > > > > To unsubscribe from this group, send email to
> > > > > > > > [email protected]
> > > > > > > > (mailto:[email protected]).
> > > > > > > > For more options, visit this group at
> > > > > > > > http://groups.google.com/group/thinking-sphinx?hl=en.
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > You received this message because you are subscribed to the
> > > > > > > Google Groups "Thinking Sphinx" group.
> > > > > > > To view this discussion on the web visit
> > > > > > > https://groups.google.com/d/msg/thinking-sphinx/-/xlq7lE__VGcJ.
> > > > > > > To post to this group, send email to
> > > > > > > [email protected]
> > > > > > > (mailto:[email protected]).
> > > > > > > To unsubscribe from this group, send email to
> > > > > > > [email protected]
> > > > > > > (mailto:[email protected]).
> > > > > > > For more options, visit this group at
> > > > > > > http://groups.google.com/group/thinking-sphinx?hl=en.
> > > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > You received this message because you are subscribed to the Google
> > > > > > Groups "Thinking Sphinx" group.
> > > > > > To view this discussion on the web visit
> > > > > > https://groups.google.com/d/msg/thinking-sphinx/-/50QMvBurEJcJ.
> > > > > > To post to this group, send email to
> > > > > > [email protected]
> > > > > > (mailto:[email protected]).
> > > > > > To unsubscribe from this group, send email to
> > > > > > [email protected]
> > > > > > (mailto:[email protected]).
> > > > > > For more options, visit this group at
> > > > > > http://groups.google.com/group/thinking-sphinx?hl=en.
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > > --
> > > > > You received this message because you are subscribed to the Google
> > > > > Groups "Thinking Sphinx" group.
> > > > > To post to this group, send email to [email protected]
> > > > > (mailto:[email protected]).
> > > > > To unsubscribe from this group, send email to
> > > > > [email protected]
> > > > > (mailto:[email protected]).
> > > > > For more options, visit this group at
> > > > > http://groups.google.com/group/thinking-sphinx?hl=en.
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > You received this message because you are subscribed to the Google
> > > > Groups "Thinking Sphinx" group.
> > > > To post to this group, send email to [email protected]
> > > > (mailto:[email protected]).
> > > > To unsubscribe from this group, send email to
> > > > [email protected]
> > > > (mailto:[email protected]).
> > > > For more options, visit this group at
> > > > http://groups.google.com/group/thinking-sphinx?hl=en.
> > > >
> > >
> > >
> > >
> > >
> > > --
> > > You received this message because you are subscribed to the Google Groups
> > > "Thinking Sphinx" group.
> > > To post to this group, send email to [email protected]
> > > (mailto:[email protected]).
> > > To unsubscribe from this group, send email to
> > > [email protected]
> > > (mailto:[email protected]).
> > > For more options, visit this group at
> > > http://groups.google.com/group/thinking-sphinx?hl=en.
> > >
> >
> >
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "Thinking Sphinx" group.
> > To post to this group, send email to [email protected]
> > (mailto:[email protected]).
> > To unsubscribe from this group, send email to
> > [email protected]
> > (mailto:[email protected]).
> > For more options, visit this group at
> > http://groups.google.com/group/thinking-sphinx?hl=en.
> >
>
>
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Thinking Sphinx" group.
> To post to this group, send email to [email protected]
> (mailto:[email protected]).
> To unsubscribe from this group, send email to
> [email protected]
> (mailto:[email protected]).
> For more options, visit this group at
> http://groups.google.com/group/thinking-sphinx?hl=en.
>
>
--
You received this message because you are subscribed to the Google Groups
"Thinking Sphinx" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/thinking-sphinx?hl=en.
