Howdy, Stranger!

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


Black Friday 2020 - NVMe and Storage deals - deploy worldwide - Page 148
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 2020 - NVMe and Storage deals - deploy worldwide

1145146148150151

Comments

  • AMS NVMe - I can't keep an active ssh for more than a few seconds. However, vnc is working fine. I will open a ticket tomorrow.

  • raindog308raindog308 Administrator, Veteran

    @hosthatch could you switch to Solus please?

    Solus is ugly and horrible but it least it works. Your panel is painful.

    • any given start, stop, mount, etc. option has a 30% chance of actually working

    • restarting a server often ignores changes

    • stopping/starting a server leaves you with a random chance the server will stay stopped

    • can't follow 302s on an ISO download

    • ISOs don't stay uploaded

    • whether a system mounts/unmounts an ISO or changes boot order is very hit-or-miss

  • DPDP Administrator, The Domain Guy

    @dosai said:
    AMS NVMe - I can't keep an active ssh for more than a few seconds. However, vnc is working fine. I will open a ticket tomorrow.

    What happened? IPv4 or IPv6?

  • @thedp said:

    @dosai said:
    AMS NVMe - I can't keep an active ssh for more than a few seconds. However, vnc is working fine. I will open a ticket tomorrow.

    What happened? IPv4 or IPv6?

    Ipv4. Network is kind of unstable, packet drops.

  • DPDP Administrator, The Domain Guy

    @dosai said:

    @thedp said:

    @dosai said:
    AMS NVMe - I can't keep an active ssh for more than a few seconds. However, vnc is working fine. I will open a ticket tomorrow.

    What happened? IPv4 or IPv6?

    Ipv4. Network is kind of unstable, packet drops.

    I'm not experiencing what you're experiencing so maybe test connecting from different locations?

  • DPDP Administrator, The Domain Guy

    @raindog308 said:
    @hosthatch could you switch to Solus please?

    Solus is ugly and horrible but it least it works. Your panel is painful.

    • any given start, stop, mount, etc. option has a 30% chance of actually working

    • restarting a server often ignores changes

    • stopping/starting a server leaves you with a random chance the server will stay stopped

    • can't follow 302s on an ISO download

    • ISOs don't stay uploaded

    • whether a system mounts/unmounts an ISO or changes boot order is very hit-or-miss

    @Emil

  • EmilEmil Member, Host Rep

    @raindog308 said:
    @hosthatch could you switch to Solus please?

    Solus is ugly and horrible but it least it works. Your panel is painful.

    • any given start, stop, mount, etc. option has a 30% chance of actually working

    • restarting a server often ignores changes

    • stopping/starting a server leaves you with a random chance the server will stay stopped

    • can't follow 302s on an ISO download

    • ISOs don't stay uploaded

    • whether a system mounts/unmounts an ISO or changes boot order is very hit-or-miss

    SolusVM API calls is a bit hit-or-miss sometimes indeed but should be way higher than 30% success rate. We are working on relying less and less on that API though as we phase away from it. Anyways I will look into our logs as what you experience doesn't sound good.

    As for the other issues, I'm afraid SolusVM access would not improve anything.

    VM staying stopped is not related to the panel, and if that happens please open a ticket and check the emergency checkbox. I have seen a few cases with an old uploaded ISO being the culprit and the ISO feature is something that we are working on at the moment. A few bugs was fixed yesterday and it will be getting better and better the coming weeks.

    I do appreciate your feedback and I am looking forward to give you a better panel experience in a near future.

  • jon617jon617 Veteran
    edited February 2021

    @raindog308 said: Solus is ugly and horrible but it least it works

    I want to be one of the few to say that I like the hosthatch interface. The simplicity helps me quickly find what I want, versus many solus sites I've used have functional issues and hard to find features, and are really ugly as you say. I have no need to upload a custom iso. I'd imagine if no working CentOS ISO image had been available, I'd be angry too, and wonder why they don't just add your ISO onto their main list of ISOs - that would be a quick fix to your issue.

    @raindog308 said: restarting a server often ignores changes
    stopping/starting a server leaves you with a random chance the server will stay stopped

    Lucky I've not had these issues. I did notice a random delay of up to a minute for each of my reboots and stops. I opened the console, press stop or reboot in a separate window, and waited until I see it take affect through the console. If I pressed reboot twice before the first reboot happened, it has done a second reboot up to a minute later. I'd wonder if I pressed reboot then stop might execute the two out of order.

  • @thedp said:

    @dosai said:

    @thedp said:

    @dosai said:
    AMS NVMe - I can't keep an active ssh for more than a few seconds. However, vnc is working fine. I will open a ticket tomorrow.

    What happened? IPv4 or IPv6?

    Ipv4. Network is kind of unstable, packet drops.

    I'm not experiencing what you're experiencing so maybe test connecting from different locations?

    Not experiencing any network issues either @dosai.

  • @jon617 said:

    @raindog308 said: Solus is ugly and horrible but it least it works

    I want to be one of the few to say that I like the hosthatch interface. The simplicity helps me quickly find what I want, versus many solus sites I've used have functional issues and hard to find features, and are really ugly as you say. I have no need to upload a custom iso. I'd imagine if no working CentOS ISO image had been available, I'd be angry too, and wonder why they don't just add your ISO onto their main list of ISOs - that would be a quick fix to your issue.

    @raindog308 said: restarting a server often ignores changes
    stopping/starting a server leaves you with a random chance the server will stay stopped

    Lucky I've not had these issues. I did notice a random delay of up to a minute for each of my reboots and stops. I opened the console, press stop or reboot in a separate window, and waited until I see it take affect through the console. If I pressed reboot twice before the first reboot happened, it has done a second reboot up to a minute later. I'd wonder if I pressed reboot then stop might execute the two out of order.

    You're not the only one, I like hosthatch panel interface too.
    Although I do not use the panel that much at all, me personally I didn't experienced the issues @raindog308 pointed out during the time I've setup my servers except the ISO one which for me was a 50/50 chance to bug out while uploading the image. Sometimes it would work, other times it would just stop uploading the image and give me the error saying ISO already exists and had to be manually removed from the system by one of the staff members. Other than that no issues but then again I'm sort of used to give it a bit when rebooting a server through any panel.

  • Daniel15Daniel15 Veteran
    edited February 2021

    @Emil While we're talking about control panel stuff, I posted a bunch of suggestions for the control panel earlier in the thread:

    @Daniel15 said:
    @hosthatch I had some small (low priority) pieces of feedback about your control panel, but wasn't sure of the appropriate way to send them through (email, PM, ticket, comment, etc), so here they are:

    1. When I click the "start" button for a VPS, I need to refresh the page to see the "running" status and connect to VNC. It would be nice for the page to update automatically, eg using WebSockets or Server-Sent Events, or even just polling (BuyVM's custom control panel just polls for the running status of the VPS every 5 seconds, which seems like a fine approach).

    2. Show logs for when the server was booted and shut down, and any errors when booting. Sometimes when I try to start my VPS, it doesn't start properly, but on the second attempt it works fine. Would be good to have an error log to see if I was doing something wrong.

    3. Automatically assign IPv6 upon provisioning, or allow adding it through the control pane;. I know you have a lot of tickets at the moment so I felt bad opening a ticket to add IPv6 to my VPS (even after prefixing it with "[Low Priority]")

    4. "Reconfigure Networking" uses eth0 as the device name, but on newer OSes (like Debian 10) the device name is actually ens3 by default, so it actually results in a broken networking config and I have to VNC in to fix it (and it's quite painful entering my server passwords over VNC as they're long and randomly generated, unique for each server). You're probably inheriting this brokenness from SolusVM though, I guess.

    5. Under the "Bandwidth" tab, show a traffic graph. SolusVM has this so you might be able to use its API (or at least the same data source)? I can just run vnstat or netdata myself so this is super low priority.

    6. I was trying to add an IPv6 address ending in ::1, but it kept showing a generic "Something went wrong. Please try again in a few minutes." error message. I was going to open a ticket, but ::2 worked fine. Maybe this could have a better error message?

    @Daniel15 said:

    Two more after another one of my servers was provisioned today :smiley:

    1. Show the current reverse DNS without having to click the pencil, so it's easy to tell that I've already set the right hostname

    2. Allow sharing uploaded ISOs across servers in different locations rather than having to re-upload it. Or alternatively, show a list of recently used ISO URLs under the "Upload ISO" heading, so I can just select the same URL again.

    and another one after all the flash sales:

    1. Don't say an invoice was "successfully created" when it really wasn't.
    Thanked by 3Emil the_doctor ferri
  • @thedp, @dosai did your server got rebooted just now?

  • PandoGulfPandoGulf Veteran
    edited February 2021

    There is a significant improvement in Tickets response time, it takes now a day or so (this is significant as HH usual take a week/weeks). Let us hope this continues to improve.

    It looks like Emil is doing most of the heavy lifting to close tickets and resolve issues.

    Thanked by 2kalimov622 Unixfy
  • @PandoGulf said:
    There is a significant improvement in Tickets response time, it takes now a day or so (this is significant as HH usual take a week/weeks). Let us hope this continues to improve.

    It looks like Emil is doing most of the heavy lifting to close tickets and resolve issues.

    Clearly, the problem was the other guy. j/k

  • EmilEmil Member, Host Rep

    @Daniel15 said:
    @Emil While we're talking about control panel stuff, I posted a bunch of suggestions for the control panel earlier in the thread:

    Thanks for the detailed feedback! As soon as some more urgent issues are fixed I will read through the whole thread and note down everything to make sure we prioritize accordingly.

  • the_doctorthe_doctor Member
    edited February 2021

    @Emil

    Congratulations (Grattis!) for winning bronze in the 2020 provider poll!

    https://www.lowendtalk.com/discussion/169775/let-provider-poll-2020-results

    You guys really deserve it. Hope Abdullah is feeling better soon.

  • All of my tickets have been answered and closed including one I opened today about the jumbo frames issues on the London network. Almost all of my tickets were either iso or internal network issues. I have not had any problems once a server has completed setup.

    Thanked by 3Falzo Unixfy ferri
  • @Daniel15 said: I have more complex tickets (eg. a ticket about jumbo frames not working on the Los Angeles private network

    They just fixed this today and it works great now!

    Jumbo frames help a lot! Throughput over WireGuard via their internal network went up from ~800 Mb/s to ~2.4 Gb/s when I switched the MTU from 1500 to 9000. :)

    I'm hoping HostHatch add a per-customer isolated private network at some point (like what BuyVM has) so that I don't need to encrypt internal traffic.

  • Hi folks. Long time now see. I'm looking for a NVMe VPS in the Netherlands or Sweden. If someone has an unnecessary NVMe VPS in these locations, PM me please :)

  • Daniel15Daniel15 Veteran
    edited February 2021

    Has anyone noticed that connectivity to Chicago is better than LA? I live in northern California so I'm physically closer to the LA server (18ms ping to LA vs 58ms ping to Chicago), yet often I get better download speeds from my Chicago VPS. I use Comcast, and for me, LA routes via Comcast -> Seabone/Sparkle -> Psychz whereas Chicago routes via Comcast -> NTT -> M247.

    @RedSox said:
    Hi folks. Long time now see. I'm looking for a NVMe VPS in the Netherlands or Sweden. If someone has an unnecessary NVMe VPS in these locations, PM me please :)

    I've got one in Amsterdam I'm not using (1 core 25% dedicated, 1 GB RAM, 10 GB space, 1TB monthly transfer), but the issue is that it was part of one of the bundles, so I'd lose the bundle discount if I transfer it. I'd need to work out what the regular pricing would be before splitting the bundle.

  • @Daniel15 said:
    Has anyone noticed that connectivity to Chicago is better than LA? I live in northern California so I'm physically closer to the LA server (18ms ping to LA vs 58ms ping to Chicago), yet often I get better download speeds from my Chicago VPS. I use Comcast, and for me, LA routes via Comcast -> Seabone/Sparkle -> Psychz whereas Chicago routes via Comcast -> NTT -> M247.

    I'm in Western Canada, and I've found Chicago speeds better. Get better sending from LetBox in LA to me than to HH in LA. Just anecdotal, nothing measured properly.

    Thanked by 1letbox
  • Daniel15Daniel15 Veteran
    edited February 2021

    @TimboJones said: I'm in Western Canada, and I've found Chicago speeds better. Get better sending from LetBox in LA to me than to HH in LA. Just anecdotal, nothing measured properly.

    I wish I had gotten the 10TB storage in Chicago rather than Los Angeles when they sent out the offer via email. I thought LA would be better given its proximity to where I live, but the disks are slower, iowait is higher, processors are slightly slower (E5-2680 v2 @ 2.8 GHz in LA vs E5-2690 v2 @ 3.0GHz in Chicago), and network isn't as good in LA. Oh well, it's still a very good deal.

  • @Daniel15 said:
    Chicago routes via Comcast -> NTT -> M247.

    Typo: M247 should be Psychz
    They use Psychz in Chicago too.

    Chicago is a (relatively) new location (for both HostHatch and Psychz), and the colocation cost should be much cheaper than LA. So there's less users, more bandwidth, better hardware thus faster speed.

    Thanked by 1the_doctor
  • Daniel15Daniel15 Veteran
    edited February 2021

    @Chocoweb said: Typo: M247 should be Psychz
    They use Psychz in Chicago too.

    Oops, you're right, I was mistaken. I thought they use M247 in Chicago as they do in some other locations.

    Thanked by 1the_doctor
  • jon617jon617 Veteran
    edited February 2021

    Psychz is so unreliable with bandwidth. With another provider, I was consistently getting just 5 megabit/sec to my virtual server at Psychz Dallas location, from a 1gbps connection within the same city and <3ms away. Switched to a Psychz location 1700 miles away, and I now get 300+ megabit/sec to it.

    I too noticed the Hosthatch Chicago (at Psychz) has performed quite well- hope it stays that way.

    Oh, and Hosthatch's Datacenters list is out of date or not applicable to anyone with a "sale" server. https://hosthatch.com/features
    Their Black Friday 2020 specials are hosted at Psychz, M247, and Hostcircle.

  • ChocowebChocoweb Member
    edited February 2021

    @jon617

    Psychz, M247, Hostcircle (and GleSYS in Stockholm, NFOrce in Amsterdam) are their "upstreams", which provide them network and colocation service.
    Some of them have their own datacenter (Psychz in LA and GleSYS). Others are using Equinix/Interxion/etc.
    So the server runs your VM do reside in the listed data center, there's nothing wrong.

    You can find the address of data centers here:
    Psychz: https://www.psychz.net/network.html
    M247: https://m247.com/services/host/colocation/

  • Ahh, I see. Thanks for the correction @Chocoweb

  • @Daniel15 said:
    6. I was trying to add an IPv6 address ending in ::1, but it kept showing a generic "Something went wrong. Please try again in a few minutes." error message. I was going to open a ticket, but ::2 worked fine. Maybe this could have a better error message?

    I've had this answer from customer support:

    We've had crosschecked with our system admin and ::1 is designated unusable from control panel perspective. It is infact by design and there's no plan to change of the design at the moment.

    And, when asked to share the reasoning behind this design:

    Unfortunately developers and system admin did not divulge such details.

  • EmilEmil Member, Host Rep

    As we now operate out of 14 global locations and different networks (and are consistently expanding), we are creating a separate support channel for networking related questions.

    We would appreciate if you reach out to support with the appropriate MTRs (both directions) if you are not satisfied with the performance. If you see better performance/latency with another provider, sending comparative results would also help a lot. We will make sure you get similar, or better, performance from our network.

    @jon617 said:
    Oh, and Hosthatch's Datacenters list is out of date or not applicable to anyone with a "sale" server. https://hosthatch.com/features
    Their Black Friday 2020 specials are hosted at Psychz, M247, and Hostcircle.

    I just double checked https://hosthatch.com/features and it is not out of date. Our BF specials are hosted in all our regular locations as listed there.

    @Chocoweb said:
    @jon617

    Psychz, M247, Hostcircle (and GleSYS in Stockholm, NFOrce in Amsterdam) are their "upstreams", which provide them network and colocation service.

    Yes and no. We are getting colocation services directly from the data centers (e.g Equinix) in some locations. NFOrce just provide us with transit as they happen to be in the same data center.

  • Thanks for the explanation.

Sign In or Register to comment.