incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dave Fisher <dave2w...@comcast.net>
Subject Re: Migration Status Page
Date Mon, 24 Oct 2011 19:20:07 GMT

On Oct 24, 2011, at 11:08 AM, TJ Frazier wrote:

> On 10/24/2011 13:42, Dave Fisher wrote:
>> 
>> On Oct 24, 2011, at 10:39 AM, Dave Fisher wrote:
>> 
>>> I reported the current outage with extension and templates. There is a link on
the migration page.
>>> 
>> 
>> These are already back in service, it took less than 20 minutes once OSUOSL received
the request.
>> 
>> THANKS!
>> 
>> Regards,
>> Dave
>> 
> Dave,
> 
> FYI, there is apparently a bug in 3.3 (and earlier?) which is contributing to beating
on that server. It seems that OO.o checks for extension updates every time it's loaded. with
Quickstarter, that's once per re-boot; without it, every time a "first" document is opened.
AFAIK, the Extension Manager was paying no attention to the schedule setting for product-update
checking, as set on the Options page.
> 
> In 3.4 Beta, I don't see this :-) but I do see a crash. :-(
> That suggests that some work was done in this area, which introduced a bug. At least
for me, writing a good issue is hard, and takes time. I will file one on this eventually.
> -- 
> /tj/

I forwarded your comment to Lance Albertson at OSUOSL - you can reach him through the email
I have quoted elsewhere.

>> Dave: If this is the case then this should be the most common URI requested
>> from that server. Is this something that you could provide? And, if
>> so, is it is possible to block or give an canned response.
>> 
>> Just curious, don't spend any time on this, I just want to know if a
>> recourse is possible.
>> 

> Lance:
> 
> If this is the URL I'm thinking of we've known about it for years. Its
> actually being hosted on our FTP infrastructure because it was getting
> so much traffic that it couldn't be hosted on the oooapps/oooapps2
> infrastructure. Lately I've had to also disable apache logs on the vhost
> because it was filling up our /var volume so fast and we didn't have
> enough space to do that.
> 
> You might verify the URL to me to be sure. If its not and still pointed
> at these two machines we may be able to force varnish to keep it cached.
> 
> -- 
> Lance Albertson
> Systems Administrator / Architect
> OSU Open Source Lab

I am thinking that we should distinguish between websites for general use by humans and web
services that are expected by the application.

This is going down frequently, is it a self-inflicted DDOS? Do any forks or down-streams also
query extensions.s.oo.o and/or templates.s.oo.o? 

Regards,
Dave
Mime
View raw message