Howdy, Stranger!

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


Black Friday 2021 - NVMe and Storage deals - Deploy in 16 global locations (APAC/EU/US) - Page 47
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.

Black Friday 2021 - NVMe and Storage deals - Deploy in 16 global locations (APAC/EU/US)

1444547495055

Comments

  • zongyouxiaozongyouxiao Member
    edited January 2022

    The server network works after they assigned a new ip for me, finally got a reply after more than a week of creating a ticket, I don't understand why it takes so long to response, the wait time is not an issue for me, I just don't like the no reply support communication style.

  • @TimboJones said:

    @v3ng said:
    I've done that like 5 times already, still the same issue

    I see that you eventually found out you were assigned a duplicate IP

    A VPS I got last year had the gateway IP assigned as its IP address! Of course, it didn't work and I had to wait for a new IP to be assigned.

    Thanked by 1lentro
  • Daniel15Daniel15 Veteran
    edited January 2022

    @govac8 said:

    @cybertech said: not that good but still within HDD specs.

    Here is a result from 12/31/2021:

    Basic System Information:
    ---------------------------------
    Processor  : AMD EPYC 7551P 32-Core Processor
    CPU cores  : 2 @ 1999.984 MHz
    AES-NI     : ✔️ Enabled
    VM-x/AMD-V : ✔️ Enabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 1.8 TiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 7.90 MB/s     (1.9k) | 103.53 MB/s   (1.6k)
    Write      | 7.93 MB/s     (1.9k) | 104.08 MB/s   (1.6k)
    Total      | 15.83 MB/s    (3.9k) | 207.61 MB/s   (3.2k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 305.31 MB/s    (596) | 431.27 MB/s    (421)
    Write      | 321.53 MB/s    (628) | 459.99 MB/s    (449)
    Total      | 626.85 MB/s   (1.2k) | 891.26 MB/s    (870)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed     
                    |                           |                 |                
    Clouvider       | London, UK (10G)          | 1.25 Gbits/sec  | 5.72 Gbits/sec 
    Online.net      | Paris, FR (10G)           | 3.37 Gbits/sec  | 3.83 Gbits/sec 
    Clouvider       | NYC, NY, US (10G)         | 713 Mbits/sec   | 1.94 Gbits/sec 
    
    Geekbench 5 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 697                           
    Multi Core      | 1339                          
    Full Test       | https://browser.geekbench.com/v5/cpu/11909342
    

    And for now:

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2021-12-28                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Mon Jan 10 14:17:59 UTC 2022
    
    Basic System Information:
    ---------------------------------
    Processor  : AMD EPYC 7551P 32-Core Processor
    CPU cores  : 2 @ 1999.984 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 1.8 TiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 1.68 MB/s      (422) | 12.73 MB/s     (198)
    Write      | 1.71 MB/s      (427) | 13.27 MB/s     (207)
    Total      | 3.40 MB/s      (849) | 26.00 MB/s     (405)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 62.58 MB/s     (122) | 78.06 MB/s      (76)
    Write      | 65.60 MB/s     (128) | 83.26 MB/s      (81)
    Total      | 128.18 MB/s    (250) | 161.33 MB/s    (157)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed     
                    |                           |                 |                
    Clouvider       | London, UK (10G)          | 933 Mbits/sec   | 3.84 Gbits/sec 
    Online.net      | Paris, FR (10G)           | 2.31 Gbits/sec  | 2.36 Gbits/sec 
    WorldStream     | The Netherlands (10G)     | busy            | 1.31 Gbits/sec 
    WebHorizon      | Singapore (400M)          | 124 Mbits/sec   | 45.3 Mbits/sec 
    Clouvider       | NYC, NY, US (10G)         | 543 Mbits/sec   | 1.17 Gbits/sec 
    Velocity Online | Tallahassee, FL, US (10G) | 862 Mbits/sec   | 1.38 Gbits/sec 
    Clouvider       | Los Angeles, CA, US (10G) | 515 Mbits/sec   | 1.03 Gbits/sec 
    Iveloz Telecom  | Sao Paulo, BR (2G)        | 104 Mbits/sec   | 757 Mbits/sec  
    
    Geekbench 5 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 558                           
    Multi Core      | 1055                          
    Full Test       | https://browser.geekbench.com/v5/cpu/12093886
    

    Not sure what happened but looks like something not good :neutral:

    Your first benchmark was probably before other people on the same node started using it. Your initial disk benchmark is very good for HDD storage but it usually doesn't stay that way forever, as a lot of the speed comes from SSD caching which will have more cache misses as the node gets busier.

    Your second benchmark is still fine for HDD, and Hosthatch generally don't do anything for complaints about benchmark speeds on storage servers (since they're not designed for speed) unless it's noticeably affecting an actual, non-benchmark use case, and even then it'd need to be below RAID HDD speed to be a problem.

    I had the same issue on my BF2020 LA storage VPS and just had to get used to the disks being slower vs other servers / locations.

    Thanked by 1lentro
  • @Daniel15 said:

    @govac8 said:

    @cybertech said: not that good but still within HDD specs.

    Here is a result from 12/31/2021:

    Basic System Information:
    ---------------------------------
    Processor  : AMD EPYC 7551P 32-Core Processor
    CPU cores  : 2 @ 1999.984 MHz
    AES-NI     : ✔️ Enabled
    VM-x/AMD-V : ✔️ Enabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 1.8 TiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 7.90 MB/s     (1.9k) | 103.53 MB/s   (1.6k)
    Write      | 7.93 MB/s     (1.9k) | 104.08 MB/s   (1.6k)
    Total      | 15.83 MB/s    (3.9k) | 207.61 MB/s   (3.2k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 305.31 MB/s    (596) | 431.27 MB/s    (421)
    Write      | 321.53 MB/s    (628) | 459.99 MB/s    (449)
    Total      | 626.85 MB/s   (1.2k) | 891.26 MB/s    (870)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed     
                    |                           |                 |                
    Clouvider       | London, UK (10G)          | 1.25 Gbits/sec  | 5.72 Gbits/sec 
    Online.net      | Paris, FR (10G)           | 3.37 Gbits/sec  | 3.83 Gbits/sec 
    Clouvider       | NYC, NY, US (10G)         | 713 Mbits/sec   | 1.94 Gbits/sec 
    
    Geekbench 5 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 697                           
    Multi Core      | 1339                          
    Full Test       | https://browser.geekbench.com/v5/cpu/11909342
    

    And for now:

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2021-12-28                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Mon Jan 10 14:17:59 UTC 2022
    
    Basic System Information:
    ---------------------------------
    Processor  : AMD EPYC 7551P 32-Core Processor
    CPU cores  : 2 @ 1999.984 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 1.8 TiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 1.68 MB/s      (422) | 12.73 MB/s     (198)
    Write      | 1.71 MB/s      (427) | 13.27 MB/s     (207)
    Total      | 3.40 MB/s      (849) | 26.00 MB/s     (405)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 62.58 MB/s     (122) | 78.06 MB/s      (76)
    Write      | 65.60 MB/s     (128) | 83.26 MB/s      (81)
    Total      | 128.18 MB/s    (250) | 161.33 MB/s    (157)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed     
                    |                           |                 |                
    Clouvider       | London, UK (10G)          | 933 Mbits/sec   | 3.84 Gbits/sec 
    Online.net      | Paris, FR (10G)           | 2.31 Gbits/sec  | 2.36 Gbits/sec 
    WorldStream     | The Netherlands (10G)     | busy            | 1.31 Gbits/sec 
    WebHorizon      | Singapore (400M)          | 124 Mbits/sec   | 45.3 Mbits/sec 
    Clouvider       | NYC, NY, US (10G)         | 543 Mbits/sec   | 1.17 Gbits/sec 
    Velocity Online | Tallahassee, FL, US (10G) | 862 Mbits/sec   | 1.38 Gbits/sec 
    Clouvider       | Los Angeles, CA, US (10G) | 515 Mbits/sec   | 1.03 Gbits/sec 
    Iveloz Telecom  | Sao Paulo, BR (2G)        | 104 Mbits/sec   | 757 Mbits/sec  
    
    Geekbench 5 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 558                           
    Multi Core      | 1055                          
    Full Test       | https://browser.geekbench.com/v5/cpu/12093886
    

    Not sure what happened but looks like something not good :neutral:

    Your first benchmark was probably before other people on the same node started using it. Your initial disk benchmark is very good for HDD storage but it usually doesn't stay that way forever, as a lot of the speed comes from SSD caching which will have more cache misses as the node gets busier.

    Your second benchmark is still fine for HDD, and Hosthatch generally don't do anything for complaints about benchmark speeds on storage servers (since they're not designed for speed) unless it's noticeably affecting an actual, non-benchmark use case, and even then it'd need to be below RAID HDD speed to be a problem.

    I had the same issue on my BF2020 LA storage VPS and just had to get used to the disks being slower vs other servers / locations.

    I am uncertain how yabs is useful for storage VPS in real world. In my case, I have mounted it as SFTP with Rai Drive and used Cryptomator to transfer my Data. I have not been able to get more than 10Mbps on home connection. Perhaps SFTP is just crazy inefficient.

  • Woo-hoo, got my CHA 10TB VPS. Have funny rounding with memory and storage but so far so good.

    rsync over ssh over wireguard is going decent (fluctuating ~250-400Mb/s -- 31-50MB/s) ... may have to try benching that w/o wireguard in between to see if I get better as the pair of ssh+wireguard does not play well with maximizing throughput if CPU is an issue.

    Thanked by 1chocolateshirt
  • NyaNya Member

    why ssh over wireguard

  • @Nya said:
    why ssh over wireguard

    Privacy oriented.

  • pbxpbx Member

    @chocolateshirt said: Privacy oriented.

    If you trust more wireguard than ssh you might consider it's more secure, but how come would it bring more privacy? People looking at what's happening on the wire will see encrypted traffic in both cases, even though the protocol won't be the same, isn't it?

  • Hm... no transfer offer so far. Something is not right, people were complaining, a lot.

  • Looks pretty good for a storage server.

    curl -sL yabs.sh | bash -s -- -ig
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2021-12-28                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Tue 11 Jan 2022 10:05:08 AM CET
    
    Basic System Information:
    ---------------------------------
    Processor  : AMD EPYC 7551P 32-Core Processor
    CPU cores  : 1 @ 1999.984 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 976.4 MiB
    Swap       : 4.0 GiB
    Disk       : 915.3 GiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ----
    Read       | 2.64 MB/s      (661) | 26.03 MB/s     (406)
    Write      | 2.65 MB/s      (664) | 26.46 MB/s     (413)
    Total      | 5.30 MB/s     (1.3k) | 52.50 MB/s     (819)
               |                      |
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ----
    Read       | 106.27 MB/s    (207) | 147.71 MB/s    (144)
    Write      | 111.91 MB/s    (218) | 157.54 MB/s    (153)
    Total      | 218.19 MB/s    (425) | 305.26 MB/s    (297)
    
    Thanked by 1chocolateshirt
  • mwtmwt Member

    Anyone else waiting for Chicago storage?

  • @mwt said:
    Anyone else waiting for Chicago storage?

    I just got my Chicago storage provisioned yesterday, I'm assuming yours will be provisioned soon.

  • still waiting for Vienna compute and a Amsterdam flash sale

  • @mwt said:
    Anyone else waiting for Chicago storage?

    Gave up on mine and asked for a refund. All I was going to use it for was backups anyway, so I went directly to rsync.net and immediately had an account that I could use to backup my data with no waiting. More expensive, but this is the way it should've been with HostHatch.

  • mwtmwt Member

    @Trav said:

    @mwt said:
    Anyone else waiting for Chicago storage?

    I just got my Chicago storage provisioned yesterday, I'm assuming yours will be provisioned soon.

    Just got provisioned now :smiley: strange coincidence.

    Thanked by 1Trav
  • @Daniel15 said:

    @TimboJones said:

    @v3ng said:
    I've done that like 5 times already, still the same issue

    I see that you eventually found out you were assigned a duplicate IP

    A VPS I got last year had the gateway IP assigned as its IP address! Of course, it didn't work and I had to wait for a new IP to be assigned.

    .1 and. 254's will be dead giveaways. Sounds like their automation ran out of cycles.

  • @xetsys said:
    Perhaps SFTP is just crazy inefficient.

    There's overhead and FTP is generally slow unless there's propriety parallel sessions or something to improve TCP throughput.

    Thanked by 1xetsys
  • 2bluesc2bluesc Member
    edited January 2022

    Received my Chicago Storage VPS server yesterday and had issues with logging in (ssh keys didn't work, root password didn't work) but blew it away and replaced it relatively effortlessly with Arch Linux and haven't looked back.

    All seemed well until I attempted to setup a WireGuard VPN between my other servers (Vultr, Digital Ocean, AT&T) and it only manages to connect once in a long while. Disabled my firewalls, netcat fails all the same, tcpdump shows the packets are never arriving. The other machines all work fine on the VPN and with the netcat trial.

    Has anyone else experienced issues with UDP traffic not making it across the HostHatch network?

  • @2bluesc said:
    Has anyone else experienced issues with UDP traffic not making it across the HostHatch network?

    I have four services in Chicago, and two each in LA, London, and Stockholm. They are all connected via wireguard and I’ve never had an issue. I have other services with different providers on the same wireguard network with no issues either.

    Thanked by 2skorous 2bluesc
  • @cablepick thanks for sharing, good know there isn't some unwritten policy I'm getting burne dby.

    The UDP issues resolved themselves for the second time last night. It's very intermittent and suspect it'll happen again unless HostHatch resolves issues on their network.

  • @chocolateshirt said:

    @Nya said:
    why ssh over wireguard

    Privacy oriented.

    Wireguard ==> I can block all connections to the server and only allow in traffic over the UDP port for wireguard which only responds if the packets match.

    Previous network protections I've setup use fwknop to listen on UDP and open up firewall rules for SSH traffic as-needed to a specific IP.

    Reasons: I was so annoyed with SSH failed auth logs and trying to parse out the noise that it seemed easiest to make it impossible to try SSH or any other service unless explicitly allowed.

    Thanked by 1nobizzle
  • @2bluesc said:
    Has anyone else experienced issues with UDP traffic not making it across the HostHatch network?

    I've also had UDP traffic issues pop up out of the blue for a few hours recently. Things started out fine. (Chicago instance for context, too)

  • @2bluesc said:
    Received my Chicago Storage VPS server yesterday and had issues with logging in (ssh keys didn't work, root password didn't work) but blew it away and replaced it relatively effortlessly with Arch Linux and haven't looked back.

    All seemed well until I attempted to setup a WireGuard VPN between my other servers (Vultr, Digital Ocean, AT&T) and it only manages to connect once in a long while. Disabled my firewalls, netcat fails all the same, tcpdump shows the packets are never arriving. The other machines all work fine on the VPN and with the netcat trial.

    Has anyone else experienced issues with UDP traffic not making it across the HostHatch network?

    Power down your server and check if your IP got duplicate assigned to someone else.

    Thanked by 2lentro 2bluesc
  • @TimboJones said:

    @2bluesc said:
    Received my Chicago Storage VPS server yesterday and had issues with logging in (ssh keys didn't work, root password didn't work) but blew it away and replaced it relatively effortlessly with Arch Linux and haven't looked back.

    All seemed well until I attempted to setup a WireGuard VPN between my other servers (Vultr, Digital Ocean, AT&T) and it only manages to connect once in a long while. Disabled my firewalls, netcat fails all the same, tcpdump shows the packets are never arriving. The other machines all work fine on the VPN and with the netcat trial.

    Has anyone else experienced issues with UDP traffic not making it across the HostHatch network?

    Power down your server and check if your IP got duplicate assigned to someone else.

    I'll try that next time assuming you're referring to the UDP issue and not the initial config. I definitely rebooted the server (at least twice) and my TCP connections had no issues uploading 1TB+.

  • @harningt said:

    Wireguard ==> I can block all connections to the server and only allow in traffic over the UDP port for wireguard which only responds if the packets match.

    I totally feel that one!

  • my gut feeling tells me some makeup flash sales coming soon

  • @cybertech said:
    my gut feeling tells me some makeup flash sales coming soon

    why? seems like they have a lot to do still

  • Still waiting APAC orders, early next week is current eta, (at this point in time)

  • @2bluesc said:

    @TimboJones said:

    @2bluesc said:
    Received my Chicago Storage VPS server yesterday and had issues with logging in (ssh keys didn't work, root password didn't work) but blew it away and replaced it relatively effortlessly with Arch Linux and haven't looked back.

    All seemed well until I attempted to setup a WireGuard VPN between my other servers (Vultr, Digital Ocean, AT&T) and it only manages to connect once in a long while. Disabled my firewalls, netcat fails all the same, tcpdump shows the packets are never arriving. The other machines all work fine on the VPN and with the netcat trial.

    Has anyone else experienced issues with UDP traffic not making it across the HostHatch network?

    Power down your server and check if your IP got duplicate assigned to someone else.

    I'll try that next time assuming you're referring to the UDP issue and not the initial config. I definitely rebooted the server (at least twice) and my TCP connections had no issues uploading 1TB+.

    You don't know when the other IP could be in use, they could have been confused AF and powered off or reinstalled, giving you normal operation intermittently.

  • add_iTadd_iT Member
    edited January 2022

    Got email about downtime in SG

    Hi,
    
    As you may have noticed, your VM based in Singapore was recently offline but came back up online as of 03:45 local time (GMT+8). 
    
    This outage was caused by human error by the data center remote hands that was conducting work in the rack to deploy new hardware for us. They unplugged the node that your VM is being hosted on by mistake and despite submitting emergency requests it took a few hours for them to react and get the node back online.
    
    We are truly sorry for the inconvenience caused by this outage. While this is a rare event we will of course investigate this incident more deeply and make sure to do all we can to prevent this from happening again in the future.
    
    Please let us know if there's anything we can assist you with.
    
    Best Regards,
    HostHatch Support Team

    Now server back to online but IPv6 is down / unreachable

    Already open ticket about this

    Let see how much time ticket will answer

This discussion has been closed.