Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


GreenValueHost: Disk I/O is Poor! Post Your dd Result Here! - Page 2
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.

GreenValueHost: Disk I/O is Poor! Post Your dd Result Here!

2

Comments

  • There are IO abusers lurking around on our nodes running massive torrent applications, ToR, etc, it's become a daily issue for us and the worst part is that we have to track these users down manually. A few things on how we're going to work on stabilizing IO:

    1. We're going to deploy a custom built script to automatically detect and suspend IO abusers

    2. We're going to be deploying new fully owned servers and they'll be something like 12 bay servers with Dual Gbit uplinks also with LSI RAID and possibly cadecache.

    There's a lot on our plate, I know, but only time will show. That's all I can say for now on this matter but we'll continue what we're doing now and monitor our nodes for abuse as frequently as possible.

    Thanks everyone.

  • Mahfuz_SS_EHLMahfuz_SS_EHL Host Rep, Veteran

    @GreenValueHost said:
    There are IO abusers lurking around on our nodes running massive torrent applications, ToR, etc, it's become a daily issue for us and the worst part is that we have to track these users down manually. A few things on how we're going to work on stabilizing IO:

    1. We're going to deploy a custom built script to automatically detect and suspend IO abusers

    2. We're going to be deploying new fully owned servers and they'll be something like 12 bay servers with Dual Gbit uplinks also with LSI RAID and possibly cadecache.

    There's a lot on our plate, I know, but only time will show. That's all I can say for now on this matter but we'll continue what we're doing now and monitor our nodes for abuse as frequently as possible.

    Thanks everyone.

    Oh, great if you can fulfil what you said. Isn't there any Statistics or Real time graph for the I/O usage for All VPS running on the node? Or any script that does so! You can easily track them with the help of that.

  • @ehostlab said:
    Oh, great if you can fulfil what you said. Isn't there any Statistics or Real time graph for the I/O usage for All VPS running on the node? Or any script that does so! You can easily track them with the help of that.

    iotop is helpful but not helpful enough. We'll get something custom tailored to our needs in order to maintain IO stability.

  • iSkyiSky Member

    @ehostlab you see that ? hey @tony20 you may read this, this is what i keep saying..... Master of excuse

    Thanked by 1Tony20
  • @iSky said:
    ehostlab you see that ? hey tony20 you may read this, this is what i keep saying..... Master of excuse

  • @ehostlab said:
    Hi,
    I got a OpenVZ Box with 4 TB (!) Ram [Although, it should be 4 GB but it shows like this], 290 GB SSD-Accelerated Disk (!) [Totally Misleading, there's no touch of SSD, neither SSD nor SSD-Caching], Maybe 2 IPv4, 4 Shared Cores of E3-1240 v2, 30 TB B/W at Texas (via CC).

    Today, when I ran dd command, I got it's 36.5 mbps which is never acceptable on Jon's so called "SSD-Accelerated" Disk Space. I'm on TX2 Node.

    Would you mind sharing your DD result here as comment, also the Name of the Node you are hosted on.

    Only Those who are using VPS from GreenValueHost

    For those who can't remember dd command, it's:

    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync

    Thanks All & Sorry for My Poor English -_-

    Said What?? something is really sounding "Fishy" and it is not my porn links site!!

    http://lowendtalk.com/discussion/24081/greenvaluehost-com-already-giving-me-a-headache#latest

  • Mahfuz_SS_EHLMahfuz_SS_EHL Host Rep, Veteran

    Not that excuse exactly. They expanded to more locations as they promised, they offer services out of CC now as they promised. Only one thing that hasn't been changed yet, is the hidden overselling policy. Either confess about your Overselling ratio or increase price in order to maintain stability.

  • iSkyiSky Member

    the part of another location is true, i believe when they said it. just the case that the I/O and hidden policies become problem

  • Mahfuz_SS_EHLMahfuz_SS_EHL Host Rep, Veteran

    @Tony20 Fishy :v

  • PcJamesyPcJamesy Member
    edited March 2014

    I don't have any issue with the performance I'm getting for the price. GVH is in no way the high end of the lowend* market. But I needed a decent amount of storage and about 2GB of ram and their providing.

    I do personally believe I'm getting my $5 a month worth.

    -bash-4.1# dd if=/dev/zero of=test bs= 64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 10.1545 s, 106 MB/s

  • BruceBruce Member
    [root@vps1 ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 6.83579 s, 157 MB/s
    

    I haven't yet worked out if the monitoring alarms are network issues or the VM not working. not ideal but getting my money's worth I think.

  • ATHKATHK Member

    @GreenValueHost said:
    1. We're going to deploy a custom built script to automatically detect and suspend IO abusers

    Better stop running those DD commands or you might get terminated...

    Thanked by 1Ian_
  • Rob92Rob92 Member
    edited March 2014

    Three Tests:
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 33.014 s, 32.5 MB/s

    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 14.2276 s, 75.5 MB/s

    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 8.56233 s, 125 MB/s

    Weird

    Located in Chicago.

  • ChaseBroylesChaseBroyles Member
    edited March 2014
    [root@bzz ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 2984.66 s, 360 kB/s
    

    I'm on the ch1 node. It's usually not this bad.

  • Mahfuz_SS_EHLMahfuz_SS_EHL Host Rep, Veteran

    @Bruce said:
    [root@vps1 ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 6.83579 s, 157 MB/s

    I haven't yet worked out if the monitoring alarms are network issues or the VM not working. not ideal but getting my money's worth I think.

    Which Location & Node you are on??

  • Mahfuz_SS_EHLMahfuz_SS_EHL Host Rep, Veteran

    @Rob92 said:
    Three Tests:
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 33.014 s, 32.5 MB/s

    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 14.2276 s, 75.5 MB/s

    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 8.56233 s, 125 MB/s

    Weird

    Located in Chicago.

    Three Different results at the same time??

  • Mahfuz_SS_EHLMahfuz_SS_EHL Host Rep, Veteran

    @ChaseBroyles said:
    [root@bzz ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 2984.66 s, 360 kB/s

    It's usually not this bad.

    Yes, not usually bad, it's worst :3

  • Rob92Rob92 Member

    @ehostlab said:
    Three Different results at the same time??

    I ran it once every minute for 3 minutes to see results. Just tried a few more times and it varies from 140MB/s - 80ishMB/s now.

  • The reason for why ch1's IO was poor a little while ago was because there was a user on the node who decided to mass torrent like 20 illegal copyrighted files every few minutes. That user has been suspended and the IO is normal again now.

  • Rob92Rob92 Member

    Wow would you look at that... Seems to be getting 100+ MB/s on average now. I am fine with that.

  • MunMun Member

    Gj everyone you are disk dosing gvh!! Proud of you.

    Mun

    Thanked by 1tchen
  • @Mun said:
    Gj everyone you are disk dosing gvh!! Proud of you.

    Mun

    There's at least like 100 people dd spamming right now as far as I can tell.

  • BruceBruce Member

    @ehostlab said:
    Which Location & Node you are on??

    NY1

  • @GreenValueHost said:
    There's at least like 100 people dd spamming right now as far as I can tell.

    100MB/s isn't a bad value. But for "SSD" it's fairly bad.

  • ATHKATHK Member

    @GreenValueHost said:
    There's at least like 100 people dd spamming right now as far as I can tell.

    100 people on one node?

  • 16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 25.2889 s, 42.5 MB/s

    Dallas

  • @ATHK

    No, I meant across all nodes.

  • @myhken said:
    Still, why do people still get GVH servers after thousands of bad reviews here on LET? Nobody using this forum should never buy a GVH server, and if they do, don't complain about it, for 99% of all here on the forum know that GVH is a bad host.
    Paying GVH just to make more drama ;) Maybe a new business model for GVH, Drama plans, with auto membership here at LET. :D :D

    GVH is not a bad host. Actually you got more than enough for what you pay.

  • ATHKATHK Member

    @GreenValueHost said:
    No, I meant across all nodes.

    Just testing you!

    I may give these guys a go I mean I don't want to do anything extravagant such as constant DD tests..

    Do you allow private torrents? Max speed 1 MB/s for personal use only? What would your stance on public torrents be if they were also maxed to that speed and X connections?

  • @ATHK i was wondering this aswell, I currently have some linux distros seeded on them. They don't take up much bandwidth as they are well seeded.

This discussion has been closed.