New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Anyone else experiencing network issues or packet loss at Skylink DC recently?
Over the past few days, I’ve been noticing significant packet loss on my VPS hosted at Skylink DC in Kerkrade (The Netherlands). Is anyone else experiencing similar issues, or could this be isolated to my provider?
PS. I’m intentionally not naming the provider to avoid blaming them unnecessarily in case the issue is related to the data center itself.
Thanks!
Thanked by 1Kolestor
Comments
I notice this on one of my vps that has skylink as an upstream too
My provider is: *beat
I experience those already for weeks. Had a ticket to my host there and the only things were "We are working on a network upgrade towards Eygelshoven, it should be there end of February". "Early March" "Next week" and last week it was "Not yet, but soon".
In the meanwhile I've taken the node that I had in Eygelshoven offline and I"m not extensively monitoring it anymore. Will check again in a few weeks to see how it's behaving. Packet loss is quite normal there, and I've not yet experienced a period longer as 3 days without complete downtime (no network connectivity or something else happening) - and that's since July last year.
BTW: I'm monitoring to/from at least 5 other DCs in the Netherlands, so it's not an n=1 thing
Awful headphones but surprisingly great servers
Kuroit, which is at skylink, also does
Here's the HetrixTools history for March and April 2025 for one VPS I have with one provider operating out of the Skylink DC:
Only 3 of those were intentional reboots of the VPS after applying updates, all the others were some network issue at the DC.
Their network status page does not show any issue: https://status.skylink-data-center.nl/
Ofcourse not.
And by "coincidence" there was an update about the network upgrade (that it's not been finished yet) shortly after this thread and my reply was here. Could be a real coincidence ofcourse, but hey, give me the next lottery numbers then.
https://status.haendler.it/
I stupidly don't have smokeping running anymore, but I don't see any short outages in hetrixtools in the last month.
Thanks for the information. Looks like it’s somewhat isolated to the DC, but not entirely.
Skylink’s status page is a complete joke. Remember that big power outage a few months back? The status page showed everything was up. So yeah, I don’t trust it one bit. That’s also one of my biggest pet peeves; fake status pages.
HetrixTools doesn’t always show outages either, since it’s mostly intermittent packet loss, but Smokeping does catch it.
My hetrixtool agent was saying that there was no data received from this provider’s VPS for 15 minutes yesterday. I could login to server and ping without any issue but it was odd.
After a reboot the problem was solved but never had any similar issue from other providers
If it helps: we do host some systems for backup purposes in the SkyLink facility. We do not use SkyLink IP network as such, but rather only let them do transport to our network in Amsterdam. I have reason to believe that they use for interconnecting with AS44592 AMS<>EYL the exact same transport layer. We are observing, over a course of the last 30 days, monitored with 20 pings every 5 minutes, over this link here a 0.0% loss and no significant changes in stdev / jitter.
While I cannot state SkyLink is the cause of what you are experiencing, I can only confirm that their transport between AMS<>EYL is unaffected.
Added:
I see that the provider hinted at above is multihomed to multiple providers in NL, and uses a common IXP connection. Presumably, this is hauled back over the transport layer I mentioned above. Given that Google (the smokeping in first post goes to 8.8.8.8) would peer over said IXP, it is reasonable to assume that traffic would not traverse AS44592. It would, therefore, be safe to say the hiccups are being introduced by said provider.
Congestion issues maybe, skylink is also full and they are building new racks in a new hall to keep up with demands
The Hetrixtools outages report is very perfect, you can even see traces to verify downtimes.
We are using it for many years and never found false notifications. Key is to whitelist there server ips...
If its congestion or flaps all are issue at DC network level
Are the skylink network issues resolved?
We haven’t experienced any such issues in the SkyLink DC. We’re running on the PletX network and everything has been very stable without exceptions so far.