Hi,
There was an intermittent network outage on curacao between about
1839 and 1930 today, which appears to have been caused by a customer
being attacked. A large amount of random small packet traffic
overwhelmed the conntrack table of the server. I was on the train
during the entirety of this event and managed to end it by turning
off the customer's network temporarily.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
At approximately 1146Z, dunkel and faustino stopped responding.
Other infrastructure in the same rack is unavailable also, and
BitFolk's colo provider believes there is a power outage in the
suite. Telehouse has been contacted and I will follow up as soon as
I have more information.
Apologies for the disruption.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
2010 11:56:00 -0000
> From: dee(a)earlsoft.co.uk
....snipped
> Stupid iPod.. :)
Whoops! Sorry about that, I have never knowingly sent a message with
my iPod - my clumsy fingers and short sight makes it impractical. I do
use it to read mail when travelling and must have clicked something by
accident when fumbling at an airport.
Steve
Hi Bitfolks,
We are looking to transfer all our web stuff to a central VM, from the
reviews i've read bitfolk looks very good, do you have any capacity opening
up soon for a 300MB VM?
-Bob
HACMan <http://hacman.org.uk>
Sent from my iPod
On 24 Feb 2010, at 12:00, users-request(a)lists.bitfolk.com wrote:
> Send users mailing list submissions to
> users(a)lists.bitfolk.com
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://lists.bitfolk.com/mailman/listinfo/users
> or, via email, send a message with subject or body 'help' to
> users-request(a)lists.bitfolk.com
>
> You can reach the person managing the list at
> users-owner(a)lists.bitfolk.com
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of users digest..."
>
>
> Today's Topics:
>
> 1. Maintenance scheduled for dunkel, 1 hour, 2010-02-26 0630Z
> (Was: Re: 2010-02-22 1559Z: Unscheduled power cycle of
> server
> dunkel) (Andy Smith)
> 2. migrating mail to gmail (Duggie)
> 3. Re: migrating mail to gmail (Ian Lovingboth)
> 4. Re: migrating mail to gmail (Duggie)
> 5. Re: migrating mail to gmail (Jocke Selin)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 23 Feb 2010 16:14:19 +0000
> From: Andy Smith <andy(a)bitfolk.com>
> Subject: [bitfolk] Maintenance scheduled for dunkel, 1 hour,
> 2010-02-26 0630Z (Was: Re: 2010-02-22 1559Z: Unscheduled
> power cycle
> of server dunkel)
> To: users(a)lists.bitfolk.com
> Message-ID: <20100223161419.GK15325(a)bitfolk.com>
> Content-Type: text/plain; charset="us-ascii"
>
> Hello,
>
> On Mon, Feb 22, 2010 at 04:30:00PM +0000, Andy Smith wrote:
>> I can see no reason and nothing logged to explain this power outage.
>> It also happened before on 2010-01-12 and 2009-10-07 so I suspect a
>> faulty PSU. A loose cable also a slim possibility. I'll be
>> scheduling a visit to the colo to check and replace in the next
>> week, and a short downtime will be associated with this.
>>
>> I'll follow up soon with the planned date/time.
>
> I will be shutting this server down starting from 0630Z 2010-02-26
> i.e. this Friday morning UK time. I will attempt to do a suspend to
> disk, so it may be some minutes before your VPS is shut down.
>
> If that doesn't work, a full shutdown/boot of your VPS will be
> required. Please try to make sure your VPS does shut down and start
> up again cleanly with all the services you expect to be running. If
> you block ping, then allowing it from 212.13.194.71 would be a big
> help to me in determining whether your VPS is working or not.
>
> I intend to change the power cable (slim possibility of loose pins)
> and the PSU (small possibility of some intermittent fault) and
> generally check that all components are secure. Based on past
> results we potentially won't know for many months if that has fixed
> things, but if it doesn't then the next step is motherboard swap.
>
> I expect Friday's work to take less than 1 hour.
>
> Cheers,
> Andy
>
> --
> http://bitfolk.com/ -- No-nonsense VPS hosting
>
Hi there,
I'm considering migrating my email handling to gmail for business.
I've used gmail as a web interface to my email for several years. It
works, and the spam filtering seems to be correct in almost every
case.
My current mail handling works as follows...
I use exim4 to handle the email for 6 domains. Only 2 of the domains
are used regularly, with about 10 accounts for each. It's not a big
set-up.
I have a droplist containing 718 'To:' addresses which definitely do
not exist - I was the target of a spam attack a few years ago. The
spammer created thousands of fake addresses using my domain. I get a
lot of attempts to these addresses, but the blocklist has been very
effective in taking the load of my server. :)
I make use of some wildcard addresses. john(a)example.org,
john-foo(a)example.org and john-bar(a)example.org and
john-amazon.com(a)example.org are all accepted as valid addresses for
user 'John'. I used to use qmail and really liked the dot qmail idea.
Of the mail which gets through, I check the senders against a couple
of SBL lists (zen.spamhaus.org and list.dsbl.org). I perform a few
more checks before I pass it on to procmail. I use procmail to filter
through a whitelist and blacklist, then through ClamAssassin and
Bitfolk's SpamAssassin daemon. If it hasn't been dumped by this stage,
it is forwarded to a gmail account for ease of reading.
Does this sound like something which could be migrated to gmail? The
SBL and (clam|spam)assassin isn't a problem - gmail does this itself.
My concern is with the wildcard addresses. I know gmail supports the
'+ syntax' e.g. john+amazon.com(a)example.org, but manually converting
all of the addresses which use a hyphen, to use a plus sign would be
an impossible task. Unless you know of a viable solution?
I think that I make my email handling more difficult than necessary -
mostly borne out of that spam attack I went through
(http://en.wikipedia.org/wiki/Peter_Francis-Macrae) - but also because
of my frequent use of wildcard addresses.
Suggestions welcome.
Thanks,
Duggie
Hi,
At approximately 1559Z today, host dunkel was power cycled. This
was not planned maintenance. The customer VPSes on this host are
being restarted as I type and I am investigating the cause further.
Please accept my apologies for the disruption.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
Hello,
I have just upgraded my VM from etch to lenny, and things look good overall.
One annoyance however is that on a reboot, the clock is ~30 mins in the
future and this is only fixed by a CRON @reboot job of "ntpdate", which
adjusts this correctly (offset -1935.350400 sec). However, this large
adjustment causes dovecot to kill itself.
On boot (via console) I can see :
..
drivers/rtc/hctosys.c: unable to open rtc device (rtc0)
..
Waiting for /dev to be fully populated...done.
Setting the system clock.
Unable to set System Clock to: Thu Jan 21 15:01:46 UTC 2010 (warning).
..
Setting the system clock.
Unable to set System Clock to: Thu Jan 21 15:01:49 UTC 2010 (warning).
Cleaning up ifupdown....
...
This is using lenny kernel 2.6.26-2-xen-686.
I tried setting "xen.independent_wallclock = 1" in /etc/sysctl.conf but this
does not make a difference.
hwclock --show
returns nothing.
Any idea why the clock needs such a large adjustment on reboot?
Cheers,
--
Alastair Sherringham
Hi,
If you're running Debian testing (squeeze) then you may have noticed
complaints from the install scripts of udev saying that your kernel
still has CONFIG_DEPRECATED_SYSFS enabled and you need to turn it
off before upgrading udev.
Don't force udev to be upgraded if this happens to you. It *will*
result in an unbootable system.
You can instead switch to the upstream kernel package which now has
support for running under Xen. The package you want is
linux-image-2.6.32-trunk-686-bigmem.
Before booting though, check some other things. Firstly, be sure
that you haven't upgraded to grub 2 (the current version of the
grub-pc package in testing is grub 2). The config format is not yet
supported by pygrub which BitFolk uses, so you would end up getting
the old kernel list from your /boot/grub/menu.lst. Instead make
sure you have grub-legacy for now. Run "update-grub" to see which
kernels it finds.
Next, the upstream kernels don't need the clocksource=jiffies kernel
command line any more either. In fact it can make things break. So
check in your menu.lst that that isn't there. Do an update-grub if
you changed anything.
After booting into 2.6.32-trunk-686-bigmem you can safely upgrade
udev.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
A bit of help with finding the right command please.
What I'm after doing is a bit of video transcoding on a 4 times a year
basis, I realise this is CPU intensive and would like to be a good
neighbour to the others who are on the same physical box as me.
How do I run a command in low priority mode?
In case the flavour of linux matters I'm on Ubuntu.
