Howdy, Stranger!

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

Sign In Register

Quick Links


Categories

In this Discussion

› › My Crissic VPS went down today and solus says it's "migrating" - 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.

My Crissic VPS went down today and solus says it's "migrating"

2

Comments

  • jvnadrjvnadr Member

    Odd... My vps is working with almost no downtime or network lost (minor ones, only), but it looked for days like it was not connected to solus. Not, it seem it's working again in solus, even if in reality it worked all this time

    I host a backup mail server there for the last couple of weeks. Before that, I hosted there the main email server, but moved elsewere because of the alerts and concerns from several LET members about issues in their infrastructure.
    I will renew in February the yearly subscription and let there the backup mail server (it's synchronize every 10 minutes the main server). Will update if there is any major change of the service level.

  • hacktekhacktek Member

    So support asked me for a 2000 packet mtr, essentially saying they won't even investigate without this. That's retarted, 100 packets are enough to tell that these levels of packet loss hit service performance hard.

    Things have gone back to normal now but it's just starting to become thing after thing after thing and support has been horrible.

  • rpollestadrpollestad Member
    edited January 2016

    Things in Miami aren't looking good.

    SSH takes 30-60 seconds to connect. Getting a shell takes another 90+ seconds.

    13:02:51 up 1 day, 14:35, 2 users, load average: 4.50, 4.90, 4.62

    My "dd" test has been running for more than 25 minutes and is only halfway done:

    590278656 Jan 23 13:03 test

    Pretty bad for a mostly idle VPS that used to run great.

    Edit: Took 35 minutes to run

    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, 2085.28 s, 515 kB/s
  • DreamCasterDreamCaster Member

    same here. this is the third time my VPS is down in last week. Things are going downhill

  • noisycodenoisycode Member
    edited January 2016

    @DreamCaster said:
    same here. this is the third time my VPS is down in last week. Things are going downhill

    IO of Boxes in both LA and Miami are terrible. Did they shrink the count of servers and bring everybody on a same boat? Would that be some strategy of QN to persuade us to leave these dirty cheap VPSes?

  • blackblack Member

    noisycode said: IO of Boxes in both LA and Miami are terrible. Did they shrink the count of servers and bring everybody on a same boat? Would that be some strategy of QN to persuade us to leave these dirty cheap VPSes?

    Judging by the latest benchmarks and how frequent downtimes occur I'd imagine so but only people at QuadraNet would know for sure.

  • namhuynamhuy Member

    MIAOVZ03

    slow network speed & slow IO

    Location                Provider        Speed
CDN                     Cachefly        30.0MB/s

Atlanta, GA, US         Coloat          4.00MB/s
Dallas, TX, US          Softlayer       62.9MB/s
Seattle, WA, US         Softlayer       13.8MB/s
San Jose, CA, US        Softlayer       6.02MB/s
Washington, DC, US      Softlayer       7.86MB/s

Tokyo, Japan            Linode          4.18MB/s
Singapore               Softlayer       2.00MB/s

Rotterdam, Netherlands  id3.net         1.02MB/s
Haarlem, Netherlands    Leaseweb        6.88MB/s


Disk Speed
----------
I/O (1st run)   : 8.7 MB/s
I/O (2nd run)   : 8.2 MB/s
I/O (3rd run)   : 5.3 MB/s
Average I/O     : 7.4 MB/s
  • daviddavid Member

    ioping Crissic/Quadranet Los Angeles:

    10 requests completed in 16.2 s, 1 iops, 5.5 KiB/s
min/avg/max/mdev = 31.1 ms / 721.9 ms / 6.2 s / 1.8 s

    ioping Crissic/Quadranet Florida:

    10 requests completed in 9.1 s, 207 iops, 830.8 KiB/s
min/avg/max/mdev = 1.1 ms / 4.8 ms / 7.3 ms / 1.8 ms

    I woke up to 76 alert emails for the server/network in Florida going down & up.

  • blackblack Member

    david said: 10 requests completed in 16.2 s, 1 iops, 5.5 KiB/s

    image

    Thanked by 1doghouch
  • DreamCasterDreamCaster Member
    edited January 2016

    noisycode said: IO of Boxes in both LA and Miami are terrible. Did they shrink the count of servers and bring everybody on a same boat? Would that be some strategy of QN to persuade us to leave these dirty cheap VPSes?

    I totally agree with you. I guess they just want Crissic clients out. Bad thing is, I had a one year prepaid 512MB Jacksonville Special and it is valid till August 2016. My VPS is on MIAOVZ01 node and 3 downtimes occurred last week.

    According to support, one happened due to a kernel panic.

    The others, weird. The VPS just stopped. I Got notificated from Uptime Robot and I booted the VPS via SolusVM. Have no idea why the VPS stopped all of a sudden. This box was working great for almost 1.5 years. I will move out once I learn how to configure this DNS thing.

    I am still here because Crissic has its own DNS network and I do not have to setup much stuff. If anyone can guide me to connect my domains to a VPS, I would be glad and leave this place.

    And this is my latest freevps bench.

    ----------
Processor       : Intel(R) Xeon(R) CPU           E5620  @ 2.40GHz
CPU Cores       : 3
Frequency       : 2400.086 MHz
Memory          : 512 MB
Swap            : 512 MB
Uptime          : 16:32,

OS              : CentOS release 6.5 (Final)
Arch            : i686 (32 Bit)
Kernel          : 2.6.32-042stab113.12
Hostname        : host.**************.com


Speedtest (IPv4 only)
---------------------
Your public IPv4 is 107.150.**********

Location                Provider        Speed
CDN                     Cachefly        94.5MB/s

Atlanta, GA, US         Coloat          77.9MB/s
Dallas, TX, US          Softlayer       59.6MB/s
Seattle, WA, US         Softlayer       27.2MB/s
San Jose, CA, US        Softlayer       31.1MB/s
Washington, DC, US      Softlayer       69.3MB/s

Tokyo, Japan            Linode          7.44MB/s
Singapore               Softlayer       7.49MB/s

Rotterdam, Netherlands  id3.net         9.20MB/s
Haarlem, Netherlands    Leaseweb        26.0MB/s


Disk Speed
----------
I/O (1st run)   : 38.2 MB/s
I/O (2nd run)   : 39.1 MB/s
I/O (3rd run)   : 51.1 MB/s
Average I/O     : 42.8 MB/s
  • ajgarettajgarett Member

    Been a weird weekend, looks like my container was moved to a new host on the 22nd, started getting alert emails about high resource usage & high io wait, SSH timing out when logging in, etc. Didn't get the host name though.

    This morning my container went down again, this time had to manually start it in SolusVM. Everything is fingers crossed working, hopefully I stick with MIAOVZ13, no more random outages/moves without notification.

  • TrafficTraffic Member
    edited January 2016

    @DreamCaster said:

    Just use your domain registrar's DNS and create the following records, with 1.2.3.4 being your VPS IP:

    Type "A" record, name "" or "@(if empty is impossible), value: 1.2.3.4

    Type "A" record, name "www", value 1.2.3.4

  • DreamCasterDreamCaster Member
    edited January 2016

    Traffic said: Just use your domain registrar's DNS and create the following records, with 1.2.3.4 being your VPS IP:

    Type "A" record, name "" or "@(if empty is impossible), value: 1.2.3.4

    Type "A" record, name "www", value 1.2.3.4

    Thank you. I have switched to my registrar's DNS and added an A record like this :

    host : www

    IP : 1.2.3.4.

    I switched to my idle XVM Labs VPS. So far everything seems to be working. I wonder if this will allow me to use the VPS'es mailing capabilities. Do I need to add MX records or anything extra?

  • sinsin Member

    DreamCaster said: I switched to my idle XVM Labs VPS. So far everything seems to be working. I wonder if this will allow me to use the VPS'es mailing capabilities. Do I need to add MX records or anything extra?

    If you just need to send mail (not receive) then you don't need to add MX records

  • ReeRee Member

    My renewal came up shortly after they got bought, and now I'm really glad I jumped ship instead of renewing!

  • DreamCasterDreamCaster Member

    sin said: If you just need to send mail (not receive) then you don't need to add MX records

    Thanks for the info. My XVM Labs VPS is definitely performing better than Crissic/QuadraNet

  • daviddavid Member

    Kicking myself now for letting my XVM Labs VPS expire. Doubtful there will ever be any new stock in the future.

    The disk problem in Los Angeles looks like it's fixed for now:

    10 requests completed in 9.0 s, 10.8 k iops, 42.2 MiB/s
min/avg/max/mdev = 67 us / 92 us / 113 us / 15 us

    Still having serious packet loss problems with the Crissic/Quadranet VPS in Florida, though (ticket open).

  • namhuynamhuy Member

    Yes I know is shared uplink, but Andrew Moore thinks "those network speeds aren't necessarily slow". @quadranet_adam @dustinc

  • amhoabamhoab Member

    @namhuy said:

    And that reply was after about five days.

  • KrisKris Member

    Sounds like slabbing.

    I've never seen OVZ able to get that granular with disk I/O.

    Probably one bare-metal machine to hold all the Crissic refugees as the brand gradually fades into the abyss, only IP assets remaining.

  • daviddavid Member

    I want to believe that Quadranet will do the right thing and not throw their new customers under the bus.

  • KrisKris Member

    Thanked by 1Francisco
  • RalliasRallias Member

    Kris said: Sounds like slabbing.

    Slabbing (at least the way that I did it) made disk I/O appear faster.

    Thanked by 1FrankZ
  • JoeMeritJoeMerit Veteran

    come on guys, just let Quadranet have their IPs!

    Thanked by 2ATHK Kris
  • KrisKris Member

    Rallias said: Slabbing (at least the way that I did it) made disk I/O appear faster.

    Not BlueVM, right? I know those were 10-20MB/s.

    I think this is more consolidate [X] OVZ nodes into [X] OVZ nodes on 1 bare-metal hypervisor.

    Just my guess, as 5.5 KiB/s is a new record for... 'sub-optimal' I/O performance.

    Makes EthernetServers seem stable. I think I can pull 15-20 MB/s on my disk there.

  • NeoonNeoon Community Contributor, Veteran

    @JoeMerit said:
    come on guys, just let Quadranet have their IPs!

  • daviddavid Member

    JoeMerit said: come on guys, just let Quadranet have their IPs!

    Hey, I paid $10 per year for each of my Crissic VPSs. That's a premium price for those IPs in a world of $6 VPSs!

  • 0xdragon0xdragon Member

    Quadranet bought Crissic with the direct knowledge that they would take the IPs and dump the customers. If I were you, stay away from them and warn others.

    This is why you should never, ever, trust providers based purely on some "reputation". Not that Quadranet ever had a good one.

  • RalliasRallias Member

    Kris said: Not BlueVM, right? I know those were 10-20MB/s.

    In fairness, previously, it was 5-10MB/s.

    If I had had my way, the move to feathur wouldn't have involved moving all of the people on one server to another effective server 1/3 as powerful.

    Thanked by 1Kris
  • DreamCasterDreamCaster Member

    JoeMerit said: come on guys, just let Quadranet have their IPs!

    I will. At the end of my billing period. Damn I prepaid a year for this :)

Sign In or Register to comment.