All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
High load during unixbench
I wish it were a joke, but we recently had a refund request because of high loads only to find that the user was running the unixbench test during the high load periods. (His normal load average was close to 0)
He complained that the load average was over 10 and was setting off his alerts and he contributed this to poor performance and wanted a refund. After trying to explain to him that unixbench runs multiple threads and would cause the load average to increase greatly and failing to get through to him, I have granted the refund.
Thought I would share my story and hopefully people realize that load average has nothing to do with performance if you are running a benchmark program that is designed to utilize all available CPU.
Comments
What a terrible provider you are! Just like that vps I bought last month. Every time I would turn it off my ping tests would fail. No network connectivity! Better believe I asked for a refund.
LOL
Yes, it is sad that pings don't get through. Don't worry, I've designed a system that drops all pings so it makes it fair.
Oh, we're terrible too. We delete your data when you reinstall your VPS! How unfair!
Pfft, we don't keep data for 40 days after a user gets terminated then wants to come back!
@Damian That's terrible.
8 CPU cores
16GB ram
1TB storage
30TB bandwidth
$1/m with promo code "lolipxcore"
Order here
LIKE A BOSS. Whatever the hell that means. SWAG YOLO? Not sure if any of these are applicable...
winning
@vdnet Why did that customer feel the need to run UnixBench all the time? How about suspending him for a change?
I'm sure that was one of his first actions... providers like to do what they can to keep money.
@marcm, it actually wasn't all the time. His load graph was small, with two large spikes where he ran unixbench. He posted a screenshot of top with the unixbench processes going to prove the load was high. That was the only reason I knew that was what was causing the load. Other than the two spikes, his load average was well below 1 the entire time.
We don't usually suspend for short bursts in CPU. If we see consistent high usage, we would warn then cap the CPU limit before resorting to suspension.
We actually had no issue with his running of unixbench, he didn't have a high average CPU. The customer opened a ticket complaining about the high loads and then requested a refund. No suspension took place.
This thread made me lol greatly, thanks! I have seen similar things but related to disk IO instead. AKA customer is running a very busy site on their SSD vps and decide to run FIO and than complain because they don't get the same numbers they had when they ran it on their fresh install with nothing running on it hehe.
Maybe he benchmarked it and didn't like what he saw so made up some bs excuse, or he just wanted to benchmark it.. People are pretty obsessed with this benchmarking and serverbear fad.
No reason why you would run it constantly though.
Being thought of as a fad makes me all giddy.
Need more BoFH http://pages.cs.wisc.edu/~ballard/bofh/bofhserver.pl
No autoboot? :P
Someone was running it on cron every hour here. I guess wanted to make a chart with results over a certain period...
M
That's bound to annoy any host, quite often on larger core cpus it takes longer than an hour to run too. I certainly don't advocate running this stuff excessively.