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.
Comments
Have three VPSes with them. Two have been down for over an hour. No reply from support yet.
Node 36 and Node 37
Currently backing up my third VPS "just in case".
@ihatetonyy I see zero there for swap, as such I doubt it's even using vSwap if the host even provided it to you. You would see an actual amount of swap space available, just like you would with xen, kvm, or a physical server if vSwap were enabled and utilized.
noooooooooooooooooooooo
I can't comment to whether it is or not but I would like to think that providers disable such features on production servers. The features are designed for power saving but in a data center environment where 10+ servers could be pushing 2A each, I don't see where the savings would come into play, especially if clients are hit with the performance reduction.
Just my 2 cents.
Well, what you are seeing is CPU scaling for power saving.
When you see this:
cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Xeon(R) CPU L5420 @ 2.50GHz
stepping : 10
cpu MHz : 1250.072
That CPU is a 2.4GHZ one. It's self underclocking 50%.
Meaning, your server is fairly idle.
It's nothing to worry about. Apply some load and the CPU comes back up to speed.
Any provider who has a server where Amp surcharge applies or power is metered on percentile use would have these enabled. Most OS'es ship with CPU scaling enabled. In general, it works fine and saves resources.
Question is with the CVPS, if you saw these clock speeds before or if it is following their big outage?
Most providers are never hands on with their server or have constant KVM over IP access to disable it from the BIOS. I've seen it on a few servers in my time.
Ideally, all providers would disable it, but these things happen. I'm sure a lot don't even know the processor has such technology, most users don't seem to know in this thread.
This is why it's good to be hands on with your hardware, or have actual access to co-located equipment. If not physical access, a KVM over IP / console and power access at least.
Seeing this out of one of the CVPS nodes:
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 42
model name : Intel(R) Xeon(R) CPU E31270 @ 3.40GHz
stepping : 7
cpu MHz : 3392.501
Guess I am not so lucky and in a server with node mates that using their resources.
That is possible. They could just have your CPU speed throttled like we do on our Backup VPS plans.
model name : Intel(R) Xeon(R) CPU 5130 @ 2.00GHz
cpu MHz : 997.510
Tickets put in at 09:43 GMT - it's now 13:19 GMT and still no reply, and both VPSes are still down.
Guess it's time to look elsewhere.
I just canceled my account there. Too much drama and unresolved ends on what happened.
Luckily I use PayPal and didn't have any card data stored with them. Folks better keep an eye on accounts if they shared their card.
@pubcrawler in theory that should not even be an issue if you've properly canceled. Just make sure you check your paypal profile for pre-approved payment authorizations or subscriptions, people normally miss those and their first instinct is to blame the provider when they themselves could have gone in and canceled the payment arrangement themselves.
I think @kbeezie CVPS in addition to Paypal accepts credit cards directly. That's who and what I am referring to.
I am fine as I don't use cards with providers since trust is inferred as sketchy at best.
After the WHMCS hack, I wouldn't trust my card info to any user of it, unless they're using a completely different payment processor for it (ie: card info/etc exists off-server with a PCI Compliant processor). Seeing as WHMCS's own setup violates PCI compliance (the card token and database exists on the same server).
Good info @kbeezie on the PCI compliance. Wasn't aware of that.
Overnight, LA node didn't go down, just dovecot stop working., other than that, o.k.
This is a joke: http://www.chicagovps.net/images/supportBox.png
5 hours. no reply to support ticket. live chat hasn't been on at all.
VPSes **still ** down on two separate nodes.
And that image has permission issues.
And honestly, give a bit longer than 5 hours for support tickets.
@HalfEatenPie out of curiosity for an emergency ticket, how long if not 5 hours would be acceptable? (except in this case the outages of several nodes appear to be widespread and known).
Quite literally, within 5 minutes of the VPS going offline I put the ticket in. Even a "yes we are working on it" would be nice.
The image was a 24/7 live support image - seems like anyone viewing an image linked from here gets a permission error; when I copy the URL into my browser it works fine.
In my opinion, a ticket marked as emergency should be looked at within an hour to establish if it is an emergency - if it's not, then it gets downgraded to normal, and put to the back of the queue. For the past five hours, I've had my client getting in touch with me asking for progress updates; which I can't give. I daren't even ask for the latest SolusVM backup I made the other day incase it pushes my emergency ticket to the bottom of the queue.
The link http://www.chicagovps.net/images/supportBox.png loaded fine for me, didn't see a permission issue, linked or copy/pasted.
@DeanClinton do you have any off-site backups handy? And by clients I assume you mean people you are in turn hosting?
Emergency tickets should be prioritized and staff should be contacted immediately.
Even small companies should have outsourced tech available for when they sleep.
For this two specific VPSes, offsite from this week, no. Last week, yes.
By clients I mean friends i'm hosting for free plus a client database for a charity I'm involved in...
I've checked their Facebook and others are down too. Someone's mentioning on Twitter that their disk has been wiped clean however I'm not sure if they are talking about today or the problem last week. I hope it's not today.
I've started migrating the data from my third VPS away from them. Yes I don't spend a fortune with them, but I think the little I do spend with them is (now was) at least worth something!
PS: I'd pickup the phone and start calling.
@pubcrawler - I was under the impression they weren't a small company per-se. However I've found out a fair bit by reading the entire thread that I never knew before.
You could always call Colocrossing for support. That's the word of who plays support now
It looks like the image is blocked based on HTTP_REFERRER
@erichi Never got that, and not sure what would be the point of wasting cycles on hot-link protection on the company page. It's probably just the way Safari opens the link that doesn't trip off the referrer protection.
@kbeezie nope, I get a 403 forbidden error.
@24khost he said referrer so that's the only info I have to go on. (403 can show up as a result of referrer protection).
yep but that is on IE so it is hot link protected.