Entered on 04/10/2013 at 12:50:51 CDT (GMT-0500) by Dmitriy Sklyar:
Closing the ticket - the server load level is a part of the new GSA implementation project.
Dmitriy.
Entered on 03/22/2013 at 11:50:53 CDT (GMT-0500) by Dmitriy Sklyar:
According to Google support, the reason our settings are not working is a bug in our version of the software (Google case #02917024).
We are in the process of implementing the new set of servers with the latest software; This should be done within next two month. I am putting this ticket on "Pending" status until implementation is complete.
Regards,
Dmitriy.
Entered on 03/06/2013 at 13:14:06 CST (GMT-0600) by raffenet@mcs.anl.gov:
I'm still seeing a very high request count coming from the 2 scanners.
root@blizzard:/var/log/apache2# grep 130.202.84.131
blizzard.access_log.2013Mar06 | wc -l
90017
root@blizzard:/var/log/apache2# grep 146.137.96.101
blizzard.access_log.2013Mar06 | wc -l
74760
150,000 page requests today alone.
Ken Raffenetti
Systems Administration Associate
MCS Division - Argonne National Laboratory
On 03/06/2013 09:11 AM, Argonne Service Desk wrote:
> [Duplicate message snipped] Entered on 03/06/2013 at 10:08:04 CST (GMT-0600) by forero@ci.uchicago.edu:
The U of C term for this is a "hull account". It's very similar to an "identity" in Userbase.
The procedure is to request the hull account from Julie Irvin (jirvin@ci.uchicago.edu). You will need to provide the following information:
Full name
Date of Birth
Contact email address
Relation to the CI (eg who are you working with, sort of sponsorship).
This will let us create a hull account. The account allows for authentication only. That means any service the person wants needs to be added individually (usually via the Grouper system).
If someone needs a access to the normal suite of privileges that come with a CNet ID, then they will also contact Julie Irvin but will need to provide more information (such as SSN and resume).
Let me know if you need more information.
David Forero
System Administrator
Computation Institute
University of Chicago Entered on 03/06/2013 at 09:11:05 CST (GMT-0600) by raffenet@mcs.anl.gov:
I don't remember if I had unblocked, then blocked them again, because
they were still blocked in my Apache config when I looked this morning.
I'm unblocking them now and will check back on the activity later today.
Ken Raffenetti
Systems Administration Associate
MCS Division - Argonne National Laboratory
On 03/04/2013 05:05 PM, Argonne Service Desk wrote:
> [Duplicate message snipped] Entered on 03/04/2013 at 17:05:08 CST (GMT-0600) by Dmitriy Sklyar:
Ken, any info on the current state of the issue? Last we’ve talked, you were going to allow access from the GSAs and see if the changes that were maid had positive effect.
Please let me know.
Thank you,
Dmitriy.
Entered on 12/05/2012 at 10:24:34 CST (GMT-0600) by Dmitriy Sklyar:
Ken has unblocked one of the servers to take at the current traffic. Further actions will be taken based on results.
Entered on 10/03/2012 at 11:42:15 CDT (GMT-0500) by David Salbego:
Dmitriy,
He provided that information.
Look at his command line cut-and-pastes.
--dave
Open Footprints Ticket: Assignees [Duplicate message snipped] Entered on 10/03/2012 at 11:39:12 CDT (GMT-0500) by Dmitriy Sklyar:
Can you please verify which GSA server these excessive scans are coming from, http://search.anl.gov/ or http://public-search.anl.gov ? Entered on 10/03/2012 at 11:17:45 CDT (GMT-0500) by Denice DiGiacomo:
Entered on 10/03/2012 at 11:03:19 CDT (GMT-0500) by raffenet@mcs.anl.gov: I unblocked to 2 scanner IPs a while back for the collab.mcs.anl.gov Confluence site. Looking at the logs from yesterday, I'm still seeing over 100,000 hits per day, per bot, so no real change.
root@blizzard:/var/log/apache2# grep 130.202.84.131 blizzard.access_log.2012Oct02 | wc -l
133292
root@blizzard:/var/log/apache2# grep 146.137.96.101 blizzard.access_log.2012Oct02 | wc -l
124296
Ken
----- Original Message -----
From: "Ken Raffenetti" <raffenet@mcs.anl.gov>
To: help@anl.gov
Sent: Thursday, September 20, 2012 9:34:14 AM
Subject: Re: INFO: Re: excessive scanning of collab.mcs.anl.gov, Ticket No. 62221, Status=Pending
I haven't gotten around to unblocking the bots. I will try to get to it
today and monitor for excessive scanning.
Ken Raffenetti
Systems Administration Associate
MCS Division - Argonne National Laboratory
On 09/17/2012 05:13 PM, Argonne Service Desk wrote:
> [Duplicate message snipped]
Entered on 09/24/2012 at 17:25:47 CDT (GMT-0500) by Dana Stasiak:
Thanks Ken.
We'll wait to hear back from you before closing this ticket.
Dana
Entered on 09/20/2012 at 09:37:12 CDT (GMT-0500) by raffenet@mcs.anl.gov:
I haven't gotten around to unblocking the bots. I will try to get to it
today and monitor for excessive scanning.
Ken Raffenetti
Systems Administration Associate
MCS Division - Argonne National Laboratory
On 09/17/2012 05:13 PM, Argonne Service Desk wrote:
> [Duplicate message snipped] Entered on 09/17/2012 at 17:13:04 CDT (GMT-0500) by Dana Stasiak:
Ken,
have you seen any more evidence of this behavior?
Dana
Entered on 07/27/2012 at 14:28:07 CDT (GMT-0500) by Dmitriy Sklyar:
I have switched both internal and external GSAs to scheduled crawl and specified minimum load for collab.mcs.anl.gov and www.alcf.anl.gov.
Please remove the blocks and let me know if the situation improves. If the blocks remain in place, no results from these sites will show up in the internal/external site search.
Entered on 07/27/2012 at 14:09:19 CDT (GMT-0500) by Malikah Reed:
assigning to web services
Entered on 07/27/2012 at 14:06:08 CDT (GMT-0500) by Dana Stasiak:
Craig, Ken,
Please accept my apologies for the difficulties you have been experiencing.
I have asked Dmitry to turn down the host load schedule for these servers. I am also creating a ticket to track this to completion.
We have been recrawling the the collection due to some changes we are making for inside Argonne.
Dana
On Jul 27, 2012, at 2:32 PM, "Craig Stacey" <stace@mcs.anl.gov> wrote:
I'd go ahead and block it until it gets figured out.
-- Craig (from my mobile)
Ken Raffenetti <raffenet@mcs.anl.gov> wrote: This scanning is still out of control. In yesterday's logs I see about
300,000 requests from each of the lab's google bots. Is there something
I should be doing to these sites to prevent this excessive scanning?
root@blizzard:/var/log/apache2# grep 146.137.96.101
blizzard.access_log.2012Jul26 | wc -l
277670
root@blizzard:/var/log/apache2# grep 130.202.84.131
blizzard.access_log.2012Jul26 | wc -l
292253
There are only about 200 pages on this entire site. Can we get support
from Google with regards to this problem? A similar issue caused me to
block the ANL google bots from www.alcf.anl.gov because it was affecting
site performance.
Ken Raffenetti
Systems Administration Associate
MCS Division - Argonne National Laboratory
On 07/10/2012 09:29 AM, Ken Raffenetti wrote:
> Google Admins,
>
> I'm seeing a pretty outrageous number of requests to
> http://collab.mcs.anl.gov. Is there any way the scanning frequency can
> be turned down on this site?
>
> In July alone, I see 1.5 million requests. This is similar to what was
> happening with www.alcf.anl.gov, which I eventually just blocked from
> being crawled by the internal Google bot.
>
> root@blizzard:/var/log/apache2# grep 146.137.96.101
> *.access_log.2012Jul* | wc -l | uniq -c
> 1 1522255
> |