savannah-hackers-public
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Savannah-hackers-public] Did the permissions on CVS lock files chan


From: Bob Proulx
Subject: Re: [Savannah-hackers-public] Did the permissions on CVS lock files change?
Date: Sun, 15 May 2016 14:19:14 -0600
User-agent: Mutt/1.5.24 (2015-08-30)

Hi Thérèse and Everyone,

Thérèse Godefroy wrote:
> Bob Proulx a écrit :
> > Note that while fencepost has an IPv6 address vcs.sv.gnu.org does
> > not.  Connections to vcs will be using IPv4.  Connections to fencepost
> > from an IPv6 enabled host will be IPv6.
> 
> That explains a few things. But my problem is that right now I can't
> connect to fencepost or to any FSF repo (Savannah or Email Self-Defense)
> in IPv4, either anonymously or via SSH.

As far as I know the only changes to fencepost, www.gnu.org,
www.fsf.org, has been the DDoS mitigation network router changes.
Only *.savannah.gnu.org made the server move and had their systems
repackaged.

(This wasn't the upcoming software upgrades that are waiting to be
enabled.  That is *our* upgrade that is happening in the future as
soon as we are given the new systems and will almost certainly result
in us breaking some things along the way.)

> > Additional the FSF implemented some type of 3rd party DDoS mitigation
> > at the router level.  Because they have been getting hit with high
> > bandwidth attacks taking not just them down but other sites near them
> > on the routers as collateral damage.
> 
> That's an important point.
> Is there a possibility that my IPv4 IP (88.177.30.2) would be
> blacklisted because I made too many unsuccessful connection attempts?

No.  The routers won't know about application level failures.  That
type of mechanism is implemented by "fail2ban" and other services.
But those are NOT causing us problems now.  However I think the DDoS
mitigation routers are *causing* application level failures.  I have
been wrong about possible problems and fixes a few times already.  But
it is one of the changes that can definitely be the cause of the
problem.

> > So many things changed last Friday the 6th that it is difficult to
> > know which of the changes is causing the current connectivity
> > problems.

It appears we have eliminted the kernel version possibility.  Ruben
returned us to the previous kernel on Friday yet problem reports
continue since then.  So not the kernel version.  Must be something
else.  I am becoming very suspicious of the DDoS mitigation.  But
perhaps it is actually a DDoS that is being mitigated?  I have no data
and therefore no real information.  Monday morning when the admins
return to work we will start up the debugging again and continue until
resolved.

> > Thank you very much for your updated report!
> 
> And thanks very much for these clues. I feel we are getting closer.  :)

I hope so!

Bob



reply via email to

[Prev in Thread] Current Thread [Next in Thread]