Moving Ask Fedora to Staging

[Rahul Sundaram]

Hi

We have been running Ask Fedora in the devel instance for a while now
and updated several times as upstream fixed bugs and responded to
feature requests

http://ask01.dev.fedoraproject.org/questions/

We have a custom CSS file thanks to Suchakra and with the help of PJP
(co-sysadmin, cc'ed),  we have configured it to run with Apache.   I
haven't setup Postfix or memcached yet on the devel instance but atleast
Postfix has been tested locally.  We will have to think about whether we
should be running our own instances or hook into the existing
infrastructure.

A SOP has been written as well

https://fedoraproject.org/wiki/Ask_Fedora_SOP

We will add more details we go forward.   As a side note, while I was
looking at memcached,  ran into a alternative python binding (pylibmc)
for memcached which apparently performs much faster and is being used by
reddit.   This has been packaged as well as the Django (django-pylibmc)
module for that.  Details at

http://amix.dk/blog/post/19471

Let me know what I need to do to move Ask Fedora to Staging

Rahul

[Paul W. Frields]

On Thu, Aug 18, 2011 at 12:17:00AM +0530, Rahul Sundaram wrote:
>  Hi
>  
>  We have been running Ask Fedora in the devel instance for a while now
>  and updated several times as upstream fixed bugs and responded to
>  feature requests
>  
>  http://ask01.dev.fedoraproject.org/questions/
>  
>  We have a custom CSS file thanks to Suchakra and with the help of PJP
>  (co-sysadmin, cc'ed),  we have configured it to run with Apache.   I
>  haven't setup Postfix or memcached yet on the devel instance but atleast
>  Postfix has been tested locally.  We will have to think about whether we
>  should be running our own instances or hook into the existing
>  infrastructure.
>  
>  A SOP has been written as well
>  
>  https://fedoraproject.org/wiki/Ask_Fedora_SOP
>  
>  We will add more details we go forward.   As a side note, while I was
>  looking at memcached,  ran into a alternative python binding (pylibmc)
>  for memcached which apparently performs much faster and is being used by
>  reddit.   This has been packaged as well as the Django (django-pylibmc)
>  module for that.  Details at
>  
>  http://amix.dk/blog/post/19471
>  
>  Let me know what I need to do to move Ask Fedora to Staging 
Should I be able to enter an answer for a question at this point?  I
tried to enter a simple answer and received a database error.

-- 
Paul W. Frields                                http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
  http://redhat.com/   -  -  -  -   http://pfrields.fedorapeople.org/
    The open source story continues to grow: http://opensource.com

[Rahul Sundaram]

On 08/18/2011 05:48 PM, Paul W. Frields wrote:
>  Should I be able to enter an answer for a question at this point?  I
>  tried to enter a simple answer and received a database error. 
We were setting it up with a new db.    Feel free to try now

http://ask01.dev.fedoraproject.org/

Rahul

[Kevin Fenzi]

On Thu, 18 Aug 2011 00:17:00 +0530
Rahul Sundaram <metherid(a)gmail.com&gt; wrote:

>  Hi
>  
>  We have been running Ask Fedora in the devel instance for a while now
>  and updated several times as upstream fixed bugs and responded to
>  feature requests
>  
>  http://ask01.dev.fedoraproject.org/questions/
>  
>  We have a custom CSS file thanks to Suchakra and with the help of PJP
>  (co-sysadmin, cc'ed),  we have configured it to run with Apache.   I
>  haven't setup Postfix or memcached yet on the devel instance but
>  atleast Postfix has been tested locally.  We will have to think about
>  whether we should be running our own instances or hook into the
>  existing infrastructure.
>  
>  A SOP has been written as well
>  
>  https://fedoraproject.org/wiki/Ask_Fedora_SOP
>  
>  We will add more details we go forward.   As a side note, while I was
>  looking at memcached,  ran into a alternative python binding (pylibmc)
>  for memcached which apparently performs much faster and is being used
>  by reddit.   This has been packaged as well as the Django
>  (django-pylibmc) module for that.  Details at
>  
>  http://amix.dk/blog/post/19471
>  
>  Let me know what I need to do to move Ask Fedora to Staging 
I think we can start looking at staging now. ;) 

Here's what I would suggest (and feedback especially from the other
application developers very welcome): 

* Add it into proxy01.stg as https://admin.fedoraproject.org/ask/
(so we can share the cookie. Or do we want to do that anymore?)

* Make a ask01.stg instance to set it up on. 

* Have it use db01.stg

* Have it use external memcached. 

With this setup, when we move to prod it would be using proxy servers,
and caching, but it's own backend. If we found that it was under too
much load, we could add a 'ask02'. That should be possible, right?
Multiple instances with a shared backend db?

This also allows us to deploy faster, as our app servers are rhel5
currently. 

I'm a bit unsure if we want the db on the ask01 instance itself, or
using a shared db backend. On the one hand thats less things in one
machine and we can reboot/restart ask01 when we might not be able to do
so to the db backend machine. But it's also another machine to back up
and manage databases on, and if we get replication working another
place we would need to replicate. 

Thoughts?

kevin

[Toshio Kuratomi]

On Thu, Aug 18, 2011 at 11:46:22AM -0600, Kevin Fenzi wrote:
>  On Thu, 18 Aug 2011 00:17:00 +0530
>  
>  I think we can start looking at staging now. ;) 
>  
>  Here's what I would suggest (and feedback especially from the other
>  application developers very welcome): 
>  
>  * Add it into proxy01.stg as https://admin.fedoraproject.org/ask/
>  (so we can share the cookie. Or do we want to do that anymore?)
>   
I think we decided this wasn't a huge loss and the benefits of going without
it were better domain name for end users and getting to use the upstream
openid auth plugin instead of maintaining our own fas auth plugin.

[snip plan I generally agree with]

>  
>  I'm a bit unsure if we want the db on the ask01 instance itself, or
>  using a shared db backend. On the one hand thats less things in one
>  machine and we can reboot/restart ask01 when we might not be able to do
>  so to the db backend machine. But it's also another machine to back up
>  and manage databases on, and if we get replication working another
>  place we would need to replicate. 
>   
If we think we're going to load balance and/or want higher availability,
then we need separate machines.  The db is always critical to keeping an
application up.  app servers for well designed apps are much less critical
if we have multiple ones.

We'd need to setup db backup/replication/etc whether its on ask01 or on
a new db server... so really those points really seem to ask whether we
should host the db on db01/db02 or make a new db server for it.

-Toshio

[Kevin Fenzi]

On Thu, 18 Aug 2011 12:40:40 -0700
Toshio Kuratomi <a.badger(a)gmail.com&gt; wrote:

>  On Thu, Aug 18, 2011 at 11:46:22AM -0600, Kevin Fenzi wrote:
>  > On Thu, 18 Aug 2011 00:17:00 +0530
>  > 
>  > I think we can start looking at staging now. ;) 
>  > 
>  > Here's what I would suggest (and feedback especially from the other
>  > application developers very welcome): 
>  > 
>  > * Add it into proxy01.stg as https://admin.fedoraproject.org/ask/
>  > (so we can share the cookie. Or do we want to do that anymore?)
>  > 
>  I think we decided this wasn't a huge loss and the benefits of going
>  without it were better domain name for end users and getting to use
>  the upstream openid auth plugin instead of maintaining our own fas
>  auth plugin. 
ok. So it will be: 

http://ask.fedoraproject.org ? 

>  [snip plan I generally agree with]
>  
>  > 
>  > I'm a bit unsure if we want the db on the ask01 instance itself, or
>  > using a shared db backend. On the one hand thats less things in one
>  > machine and we can reboot/restart ask01 when we might not be able
>  > to do so to the db backend machine. But it's also another machine
>  > to back up and manage databases on, and if we get replication
>  > working another place we would need to replicate. 
>  > 
>  If we think we're going to load balance and/or want higher
>  availability, then we need separate machines.  The db is always
>  critical to keeping an application up.  app servers for well designed
>  apps are much less critical if we have multiple ones.
>  
>  We'd need to setup db backup/replication/etc whether its on ask01 or
>  on a new db server... so really those points really seem to ask
>  whether we should host the db on db01/db02 or make a new db server
>  for it. 
ok. I kinda dislike putting more in the same db box, but thats a
problem for all the existing dbs too, that we should solve at the db
level (with replication, etc). So, we can just stick it on db01.stg /
real db. 

I assume there's no issue in the app running on multiple machines with
the same db backend. Mether? 

kevin

[Rahul Sundaram]

On 08/19/2011 09:39 PM, Kevin Fenzi wrote:
>  ok. So it will be: 
>
>  http://ask.fedoraproject.org ?  
Yes.   I think we are going to not use the FAS auth plugin and instead
use Fedora account as just a openid.  Less code to maintain and less
things to break

>  ok. I kinda dislike putting more in the same db box, but thats a
>  problem for all the existing dbs too, that we should solve at the db
>  level (with replication, etc). So, we can just stick it on db01.stg /
>  real db. 
>
>  I assume there's no issue in the app running on multiple machines with
>  the same db backend. Mether?  
I will confirm this with upstream

Rahul

[Rahul Sundaram]

On 08/19/2011 09:39 PM, Kevin Fenzi wrote:
>  I assume there's no issue in the app running on multiple machines
> with
>  the same db backend. Mether? kevin 
Not sure this is helpful but this is the upstream answer

http://askbot.org/en/question/596/multi-instances-sharing-one-db-a-proble...

Rahul

[Kevin Fenzi]

On Sun, 21 Aug 2011 20:49:17 +0530
Rahul Sundaram <metherid(a)gmail.com&gt; wrote:

>  On 08/19/2011 09:39 PM, Kevin Fenzi wrote:
>  > I assume there's no issue in the app running on multiple machines
>  > with the same db backend. Mether? kevin
>  
>  Not sure this is helpful but this is the upstream answer
>  
>  http://askbot.org/en/question/596/multi-instances-sharing-one-db-a-proble...
Yeah, not too useful there. 

What we might want to do is spin up a ask02.stg before we move it to
production just to see if it can handle multiple instances ok. 

kevin

[Toshio Kuratomi]

On Sun, Aug 21, 2011 at 08:49:17PM +0530, Rahul Sundaram wrote:
>  On 08/19/2011 09:39 PM, Kevin Fenzi wrote:
>  > I assume there's no issue in the app running on multiple machines with
>  > the same db backend. Mether? kevin
>  
>  Not sure this is helpful but this is the upstream answer
>  
>
> http://askbot.org/en/question/596/multi-instances-sharing-one-db-a-proble...
>   
I added a followup question there.

-Toshio