Hi,
I noticed that as of at least yesterday, BitFolk's DNS resolvers are
blocked from querying URIBL¹.
URIBL is a DNS-based blocklist of URIs mentioned in email spam. It
is used by default in SpamAssassin and other anti-spam products.
The usual reason for being blocked from querying URIBL is excessive
query volume. Their web site mentions a figure of 100k queries per
day.
I've had a look at the usage from BitFOlk's SpamAssassin service²
and the cluster is only checking about 7k emails per day. Possibly
7k emails expands to over 100k URIs to query, or possibly other
customers are doing excessive DNS queries - anyone using BitFolk's
resolvers to query will add to the count. I have asked URIBL for
clarification of what the issue is.
If the issue is query load then we will pay for a feed.
Until that is sorted out you will unfortunately be unable to query
URIBL usefully. It returns a value that SpamAssassin recognises as
"query blocked".
Cheers,
Andy
¹ https://uribl.com/
² https://www.bitfolk.com/customer_information.html#toc_2_SpamAssassin
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
TL;DR: Bitcoin payments are broken, hopefully fixed today or
tomorrow. If you need to make a payment that way please wait a
little bit. If your service will be terminated for non-payment
please contact support for an extension until it's fixed.
More detail:
It seems that our bitcoind is crashing, I think because I made the
mistake of running it as 32-bit (see repeated threads about the
death of 32-bit) and it now needs to allocate more memory than a
32-bit process can.
I could give it a 64-bit kernel but the bitcoind itself would still
be 32-bit so I'd have to recompile it as well. I'm instead going to
rebuild that host.
Cheers,
Andy
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
At some point between 16:00Z and 17:00Z today I'm going to upgrade
the rescue VM¹ image from Debian 9 (stretch, oldstable) to Debian 10
(buster, stable).
This will only take a few seconds, but as the rescue VM is an
NFS-mounted squashfs image, replacing the image will prevent all
currently running copies of the rescue VM from functioning.
If you are using the rescue VM at the time, you will start to see
I/O errors on its on root filesystem and will have to kill it and
start it again. At its next boot a new image based on Debian 10 will
load.
If you are or expect to be in the middle of something important with
the rescue VM between 16:00 and 17:00 then let us know and we'll
postpone the work.
Cheers,
Andy
¹ https://tools.bitfolk.com/wiki/Rescue
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
Briefly:
At around 2153Z today we started receiving alerts for basically
"everything" flapping up and down due to between 20 and 30% packet
loss. In short, this was due to a distributed denial of service
attack aimed at another customer of our colo provider. Around 2240Z
they deployed some mitigation and hopefully we don't see any more
issues.
TL;DR:
I'd only got as far as determining that it affected "everything"
before it stopped again shortly before 2200. It then started up
again around 2215 and I was able to see that it also affected our
colo provider.
I was able to make contact and they started investigating around
2215. There wasn't anything for me to do at this point except watch.
It was a large UDP DDoS, random source and destination ports.
At about 2240 they put in some mitigation. The degree to which you
were affected during this time will vary based on how your legit
traffic reached us (or didn't reach us).
Cheers,
Andy
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
I've added a warning to the Xen Shell "install" command that new
installs of 32-bit guests are now deprecated, and pointing you to:
https://tools.bitfolk.com/wiki/64-bit_guests
Those of you with existing 32-bit guests, which I know is many of
you, should be okay for a few years yet. Even after the Xen
hypervisor officially ditches support for 32-bit guests we will
ensure they are still bootable. But I would encourage you to choose
64-bit at your next install or operating system upgrade for an
easier life.
Cheers,
Andy
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
Some security issues have been found in the hypervisor software we
use, which we have to fix as they can theoretically allow privilege
escalation.
They are under embargo until Thursday 31 October, so we will most
likely do the work in the early hours of the morning (UK time) on
29, 30, and 31 October.
As usual this will entail a clean shutdown of your guest and then a
boot again 20–30 minutes later after the patching is done. Some time
next week an email will go out telling you of the two hour window in
which this work will take place for each of your VMs.
If the assigned window is unacceptable to you, we can most likely
move your VM to an already-patched host at a time of your choosing
before 31 October. When the direct email comes to let you know of
your maintenance window, if it's not acceptable then you can reply
to it to open a support ticket and we will work it out.
As usual, if you have opted in to suspend/restore then your guest
will be suspended to disk and restored again instead of shutdown and
booted. More info on that:
https://tools.bitfolk.com/wiki/Suspend_and_restore
For our own maintenance work we like to give more than 2 weeks of
notice. Unfortunate when dealing with security issues there is an
agreed embargo process and notice periods are much shorter. It is
preferable that there is ~2 weeks of notice rather than a "0-day"
exploit being unleashed.
Thanks,
Andy
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
A few customers have been testing this for a while now, and it's
been a while since the last issues were addressed, so now seems like
a good time to announce it.
We're going to be retiring our Cacti instance¹ in favour of the new
setup which can be found at:
https://tools.bitfolk.com/grafana/
You all already have access to it.
Those who are familiar with Prometheus and Grafana may be a little
disappointed: this is not intended to be a full hosted instance,
only a fairly locked-down replacement for what Cacti provides. I'm
satisfied that it goes beyond the functionality and usability of
Cacti, but it isn't like having your own setup and isn't intended to
be.
Everyone has a default dashboard exposing graphs similar to those
provided by Cacti, plus a few more besides.
The offer was always open for more of your metrics to be graphed by
Cacti, but as of today only one customer was making use of that. The
offer is still open for us to graph extra metrics from you if you
wish. To do that you'll first need to install Node Exporter² and
then send a support ticket. You'll then get an additional
dashboard that looks like a bit like this:
https://tools.bitfolk.com/grafana/dashboard/snapshot/fysbHKJGqJm3Fq6KmtqlRJ…
Over the next week or two a wiki article for our Grafana will appear
and any references to Cacti on our web sites and docs will start to
disappear, except for a pointer to historical Cacti graphs. Update
of Cacti graphs is going to be disabled very soon.
Feedback on the service is still welcome of course, though the
general approach is by now pretty much decided.
Cheers,
Andy
¹ https://tools.bitfolk.com/cacti/
² Available as your usual kind of single Go binary from here:
https://github.com/prometheus/node_exporter
but also available in modern Debian (at least) as a package.
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
Another serious bug has been found in Exim, which is installed by
default on Debian and some other Linux distributions:
https://seclists.org/oss-sec/2019/q3/253
The impact is remote execution as an unprivileged user, although
it cannot be ruled out that there might be other routes to the same
code running in a privileged context.
If your distribution is still under security support then I expect
they will push out new packages in the next few days.
If not then you will need to upgrade it or rebuild the package. It's
quite a simple fix.
There's been no embargo this time, so attacks could be out in the
wild already.
Cheers,
Andy
--
https://bitfolk.com/ -- No-nonsense VPS hosting
Hi,
TL;DR: Read this to learn how to install CentOS 8
https://tools.bitfolk.com/wiki/Installing_CentOS_8
Unabridged edition:
Given that CentOS 8 was released a few days ago I had a look at
adding its installer.
Unfortunately it seems that CentOS 8 has dropped kernel support for
PV-mode Xen guests, which are the only type of guests that BitFolk
currently supports. It is therefore not possible to use the official
CentOS installer or core kernel package at the moment.
We are in the process of moving to PVH mode¹ guests, but that's not
ready yet. It all works; the main difficulty now is supporting both
modes without it being a terribly confusing user experience.
In the meantime, it is pretty simple to install CentOS 8 from
another Linux. This could be any distribution including an earlier
version of CentOS, though I would suggest that doing it from the
BitFolk Rescue VM² makes most sense as it's always available and
runs from RAM.
As the core kernel package of CentOS 8 also does not support PV mode
guests, it is also necessary to enable ELRepo³ and install the
kernel-ml package.
Here is a transcript of me installing CentOS 8 from scratch by this
method with full explanation of every step.
https://tools.bitfolk.com/wiki/Installing_CentOS_8
Don't be put off by the massive amount of text here; the vast majority
of it is command output which I have only included so you know what
to expect.
The only issue I have found with this method are some odd 1–2 minute
pauses around creating initramfs / bootloader config. This only
happens inside the install chroot and is probably something trying
to probe and timing out. It appears to be harmless, just irritating.
If you know what that is about or have any other improvements to
make, please do edit the page⁴; it is a wiki.
Cheers,
Andy
¹ https://wiki.xen.org/wiki/Xen_Project_Software_Overview#PVH_.28x86.29
² https://tools.bitfolk.com/wiki/Rescue
³ https://elrepo.org/tiki/kernel-ml
⁴ I would suggest refraining from adding purely optional things that
are a matter of taste though, as otherwise the page will become
incredibly long and opinionated.
--
https://bitfolk.com/ -- No-nonsense VPS hosting