FastCGI

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

FastCGI

Scooter Willis-2

In looking through the apache error logs it looks like fastcgi is failing because of a timeout. Since I am trying to do a memory only version to eliminate the complexity of the database and to allow editing of gff3 files does the fastcgi implementation not work in this scenario? I will see if I can find an option to increase the timeout but not sure if future calls will have everything already cached.

[Wed Jun 30 10:19:26 2010] [warn] FastCGI: (dynamic) server "/usr/lib/cgi-bin/gb2/gbrowse" restarted (pid 2318)
[Wed Jun 30 10:19:56 2010] [error] [client 127.0.0.1] FastCGI: comm with (dynamic) server "/usr/lib/cgi-bin/gb2/gbrowse" aborted: (first read) idle timeout (30 sec)
[Wed Jun 30 10:19:56 2010] [error] [client 127.0.0.1] FastCGI: incomplete headers (0 bytes) received from server "/usr/lib/cgi-bin/gb2/gbrowse"

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Gmod-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-devel
Reply | Threaded
Open this post in threaded view
|

Re: FastCGI

Scott Cain
Hi Scooter,

I'm sending this one over to the GBrowse mailing list as well.

Scott


On Wed, Jun 30, 2010 at 10:21 AM, Scooter Willis <[hidden email]> wrote:

>
> In looking through the apache error logs it looks like fastcgi is failing because of a timeout. Since I am trying to do a memory only version to eliminate the complexity of the database and to allow editing of gff3 files does the fastcgi implementation not work in this scenario? I will see if I can find an option to increase the timeout but not sure if future calls will have everything already cached.
>
> [Wed Jun 30 10:19:26 2010] [warn] FastCGI: (dynamic) server "/usr/lib/cgi-bin/gb2/gbrowse" restarted (pid 2318)
> [Wed Jun 30 10:19:56 2010] [error] [client 127.0.0.1] FastCGI: comm with (dynamic) server "/usr/lib/cgi-bin/gb2/gbrowse" aborted: (first read) idle timeout (30 sec)
> [Wed Jun 30 10:19:56 2010] [error] [client 127.0.0.1] FastCGI: incomplete headers (0 bytes) received from server "/usr/lib/cgi-bin/gb2/gbrowse"
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Gmod-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gmod-devel
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Gmod-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-devel
Reply | Threaded
Open this post in threaded view
|

Re: FastCGI

Scooter Willis-2
With a little google searching based on the error message I added -idle-timeout 120 to gbrowse2.conf
FastCgiConfig -idle-timeout 120 -initial-env GBROWSE_CONF=/etc/gbrowse2

I now get the page to load and figured out that the fastcgi call runs as gbrowse process.

The behavior is very similar to the spawning of processes using mod_perl option.

Once the initial track is loaded I see three gbrowse processes each consuming 600 MB of memory. In my configuration I am trying to load everything into memory for our genome + 6 different gff3 tracks related to three different gene prediction algorithms.

It appears that each process is run per request versus some level of persistence where once everything is loaded into memory it can service the requests without loading from disk. Unless I am missing something it looks like I need to load everything into a database. I was really trying to avoid that complexity.

Any guidance on the overall architecture would be a big help. We have been using mod_perl option which performs well but I discovered this morning that I had lots of mod_perl processes each allocating 600MB of memory.

Thanks

Scooter

________________________________________
From: Scott Cain [[hidden email]]
Sent: Wednesday, June 30, 2010 10:57 AM
To: Scooter Willis
Cc: [hidden email]; Gbrowse (E-mail)
Subject: Re: [GMOD-devel] FastCGI

Hi Scooter,

I'm sending this one over to the GBrowse mailing list as well.

Scott


On Wed, Jun 30, 2010 at 10:21 AM, Scooter Willis <[hidden email]> wrote:

>
> In looking through the apache error logs it looks like fastcgi is failing because of a timeout. Since I am trying to do a memory only version to eliminate the complexity of the database and to allow editing of gff3 files does the fastcgi implementation not work in this scenario? I will see if I can find an option to increase the timeout but not sure if future calls will have everything already cached.
>
> [Wed Jun 30 10:19:26 2010] [warn] FastCGI: (dynamic) server "/usr/lib/cgi-bin/gb2/gbrowse" restarted (pid 2318)
> [Wed Jun 30 10:19:56 2010] [error] [client 127.0.0.1] FastCGI: comm with (dynamic) server "/usr/lib/cgi-bin/gb2/gbrowse" aborted: (first read) idle timeout (30 sec)
> [Wed Jun 30 10:19:56 2010] [error] [client 127.0.0.1] FastCGI: incomplete headers (0 bytes) received from server "/usr/lib/cgi-bin/gb2/gbrowse"
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Gmod-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gmod-devel
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Gmod-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-devel