Howdy, Stranger!

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


fastpipe.io - SSD Cloud Servers (KVM) - Frankfurt, Germany - Highly Available Cloud! - 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.

fastpipe.io - SSD Cloud Servers (KVM) - Frankfurt, Germany - Highly Available Cloud!

2»

Comments

  • Mutil Blog Network allowed? and BGP

  • jh_aurologicjh_aurologic Member, Patron Provider

    @s3wd0ugh said:
    Do you have a refund policy?

    Sure, the german "Widerrufsrecht" applies, which means that you can retract a contract as long as it has not been used or paid.

    @SnowStylez said:
    Mutil Blog Network allowed? and BGP

    You can do whatever you want as long as it does not breach our tos: https://www.fastpipe.io/tos

    BGP is free: https://www.fastpipe.io/bgpsessions

  • TheLinuxBugTheLinuxBug Member
    edited February 2020

    azbot said: I created a ticket to support 10 days ago. It was unbelievable that I was not supported yet, so I canceled the service.

    I will say this, I feel like their support quality and service quality has dropped off recently.

    A ticket I submitted got some canned reply blaming IO issues I reported on the product I purchased being an 'hdd' based product when IO to my server was <10M/sec and as far as I know, still is and hasn't been addressed. Instead they told me that I should upgrade to an SSD product instead of even reviewing my report. This is a very poor response to someone voicing an issue which likely also effects other customers on the same node.

    I plan to discontinue the server after my current service period because of this and have already moved the sites I was hosting there away to another provider.

    Additionally, the bandwidth quality has also dropped off since I started using their services, however, I have taken this is stride so far as they have reportedly been upgrading their network.

    I wouldn't suggest against them per se, but I am for sure no longer an advocate of their product after my recent experiences with their services and support.

    my 2 cents.

    Cheers!

  • jh_aurologicjh_aurologic Member, Patron Provider
    edited February 2020

    @TheLinuxBug said:

    azbot said: I created a ticket to support 10 days ago. It was unbelievable that I was not supported yet, so I canceled the service.

    I will say this, I feel like their support quality and service quality has dropped off recently.

    What sort of support do you expect from a small instance which generates around 2,50€ of revenue per month? Sure, we might have been faster as we were smaller.

    All tickets are regulary handled within 24 hours, except accounting.

    A ticket I submitted got some canned reply blaming IO issues I reported on the product I purchased being an 'hdd' based product when IO to my server was <10M/sec and as far as I know, still is and hasn't been addressed. Instead they told me that I should upgrade to an SSD product instead of even reviewing my report. This is a very poor response to someone voicing an issue which likely also effects other customers on the same node.

    You dont have a server with SSD Storage, you just have a small HDD instance from a very old offer back a few years ago. Maybe it is swapping? Node shows not even 1% IOwait. Back then all instances had swap enabled.

    Below are pictures of our Grafana showing both the Nodes load and IOwait:


    I plan to discontinue the server after my current service period because of this and have already moved the sites I was hosting there away to another provider.

    Thats bad, because there are plans to introduce HDD Cloud Storage with NVMe Caching until the end of April, which means there will be no single nodes anymore. Higher redundancy, more I/O and customizable storage.

    Additionally, the bandwidth quality has also dropped off since I started using their services, however, I have taken this is stride so far as they have reportedly been upgrading their network.

    There were a few hickups from the end of December till mid of January. We got a unexpected big customer who increased the pressure on a dc interconnection, where we had to wait around four weeks for additional capacity. That capacity is now live: https://www.fastpipe.io/blog/?p=189

  • combahton_it said: What sort of support do you expect from a small instance which generates around 2,50€ of revenue per month? Sure, we might have been faster as we were smaller.

    I would not post such statements publicly, even if there is some truth in it. It doesn't seem very professional. But hey, I've read much funnier threads with you on dev-tek. Your arrogance is remarkable.

    Thanked by 1webcraft
  • jh_aurologicjh_aurologic Member, Patron Provider
    edited February 2020

    @Tr33n said:
    I would not post such statements publicly, even if there is some truth in it. It doesn't seem very professional. But hey, I've read much funnier threads with you on dev-tek. Your arrogance is remarkable.

    I'm just directly asking. I want to know what he expects from such a small instance. Thats honest minded, service oriented and is not related to arrogance. If it would be kind of arrogant, I doubt that I would have made myself the effort to test the I/O on the node in question and share screenshots of the load.

    dev-tek - probably wasnt active there for over twelve or sixteen months. Too bad we dont know us personally, building a oppinion about someone from forum posts is prone to misunderstandings ;-)

  • I just recently bought yesterday a vps via paypa. The credit i added didnt reflect. But theres no credentials emailed yet. Ive also create a ticket but no response yet. #855947923

  • @combahton_it

    do you host fastpipe on same DC as your looking glass? I almost sold after pinging fastpipe.io but fall back when pinging lg.combahton.net ~20ms difference.

      6. AS131111 ip-103-78-83-161.moratelindo.net.id   0.0%    10   17.6  19.1  16.4  21.6   1.8
      7. AS???    103.83.6.101                          0.0%    10   16.7  17.6  16.2  19.5   1.2
      8. AS???    103.83.6.18                           0.0%    10   18.6  17.6  15.6  20.1   1.3
      9. AS???    33891.sgw.equinix.com                 0.0%    10   17.6  17.9  16.5  20.2   1.3
     10. AS201011 ae6-2090.fra20.core-backbone.com      0.0%    10  197.5 198.9 196.4 202.4   2.0
     11. AS30823  ae1-33891.edge1.ffm3                  0.0%    10  216.7 232.9 182.7 340.1  62.5
     12. AS30823  ae0-edge1.core1.ffm3                  0.0%    10  207.1 232.1 202.5 328.8  41.2
     13. AS30823  www-vip.fastpipe.io                   0.0%    10  176.3 177.3 175.3 180.0   1.7
    

    Looking glass

      6. AS131111 ip-103-78-83-161.moratelindo.net.id   0.0%    10   18.4  18.6  16.8  20.2   1.1
      7. AS???    103.83.6.101                          0.0%    10   17.9  18.4  16.5  20.5   1.4
      8. AS???    103.83.6.18                           0.0%    10   17.4  18.4  17.4  19.9   0.8
      9. AS???    33891.sgw.equinix.com                 0.0%    10   18.6  18.7  17.0  21.2   1.3
     10. AS201011 ae6-2090.fra20.core-backbone.com      0.0%    10  199.5 198.7 196.6 200.5   1.3
     11. AS30823  ae1-33891.edge1.ffm3                  0.0%    10  267.9 209.2 180.4 347.7  55.3
     12. AS30823  ae0-edge1.core1.ffm3                  0.0%    10  217.5 231.6 206.6 294.0  33.9
     13. AS30823  ae1-ffm31.core1.ffm2                  0.0%    10  205.9 200.8 181.4 254.0  20.3
     14. AS30823  lg.combahton.net                      0.0%    10  197.5 198.7 197.3 200.2   1.1
    
    
  • jh_aurologicjh_aurologic Member, Patron Provider

    @arjay316 said:
    I just recently bought yesterday a vps via paypa. The credit i added didnt reflect. But theres no credentials emailed yet. Ive also create a ticket but no response yet. #855947923

    Business hours are Mo-Fr from 09:00 am till 06:00 pm.

  • jh_aurologicjh_aurologic Member, Patron Provider

    @sibaper said:
    @combahton_it

    do you host fastpipe on same DC as your looking glass? I almost sold after pinging fastpipe.io but fall back when pinging lg.combahton.net ~20ms difference.

    >   6. AS131111 ip-103-78-83-161.moratelindo.net.id   0.0%    10   17.6  19.1  16.4  21.6   1.8
    >   7. AS???    103.83.6.101                          0.0%    10   16.7  17.6  16.2  19.5   1.2
    >   8. AS???    103.83.6.18                           0.0%    10   18.6  17.6  15.6  20.1   1.3
    >   9. AS???    33891.sgw.equinix.com                 0.0%    10   17.6  17.9  16.5  20.2   1.3
    >  10. AS201011 ae6-2090.fra20.core-backbone.com      0.0%    10  197.5 198.9 196.4 202.4   2.0
    >  11. AS30823  ae1-33891.edge1.ffm3                  0.0%    10  216.7 232.9 182.7 340.1  62.5
    >  12. AS30823  ae0-edge1.core1.ffm3                  0.0%    10  207.1 232.1 202.5 328.8  41.2
    >  13. AS30823  www-vip.fastpipe.io                   0.0%    10  176.3 177.3 175.3 180.0   1.7
    > 

    Looking glass

    >   6. AS131111 ip-103-78-83-161.moratelindo.net.id   0.0%    10   18.4  18.6  16.8  20.2   1.1
    >   7. AS???    103.83.6.101                          0.0%    10   17.9  18.4  16.5  20.5   1.4
    >   8. AS???    103.83.6.18                           0.0%    10   17.4  18.4  17.4  19.9   0.8
    >   9. AS???    33891.sgw.equinix.com                 0.0%    10   18.6  18.7  17.0  21.2   1.3
    >  10. AS201011 ae6-2090.fra20.core-backbone.com      0.0%    10  199.5 198.7 196.6 200.5   1.3
    >  11. AS30823  ae1-33891.edge1.ffm3                  0.0%    10  267.9 209.2 180.4 347.7  55.3
    >  12. AS30823  ae0-edge1.core1.ffm3                  0.0%    10  217.5 231.6 206.6 294.0  33.9
    >  13. AS30823  ae1-ffm31.core1.ffm2                  0.0%    10  205.9 200.8 181.4 254.0  20.3
    >  14. AS30823  lg.combahton.net                      0.0%    10  197.5 198.7 197.3 200.2   1.1
    > 
    > 

    Our website is located at our FFM3 location, the LG at FFM2 where our Cloud operates. Do you have the reverse path available?

  • Your looking glass, broken right now. I cant do traceroute

    traceroute to 8.8.8.8 (8.8.8.8), 30 hops max, 60 byte packets
     1  * * *
     2  * * *
    -- Traceroute timed out --
    

    but mtr working, reverse mtr

    Start: Sun Feb  9 17:30:55 2020
    HOST: lg                                      Loss%   Snt   Last   Avg  Best  Wrst StDev
    1.|-- ???                                     100.0    10    0.0   0.0   0.0   0.0   0.0
    2.|-- ???                                     100.0    10    0.0   0.0   0.0   0.0   0.0
    3.|-- te103-197071.edge1.ffm2.combahton.net    0.0%    10    2.7  18.7   2.7  30.6  10.5
    4.|-- 80.95.152.197                            0.0%    10    0.7   1.1   0.6   3.4   0.7
    5.|-- 80.95.144.27                             0.0%    10    7.5   8.8   7.3  13.2   2.1
    6.|-- TenGE13-2.br02.ams01.pccwbtn.net         0.0%    10    7.8   7.8   7.5   8.4   0.0
    7.|-- HundredGE0-1-0-0.br03.sin03.pccwbtn.net  0.0%    10  166.6 166.8 166.5 167.6   0.0
    8.|-- HundredGE0-1-0-0.br03.sin03.pccwbtn.net  0.0%    10  166.4 166.6 166.2 167.9   0.3
    9.|-- linknet.be12.br03.sin03.pccwbtn.net      0.0%    10  169.3 168.9 166.2 172.9   1.7
    10.|-- fm-dyn-111-95-244-98.fast.net.id         0.0%    10  173.0 175.9 172.9 200.8   8.7
    11.|-- ???                                     100.0    10    0.0   0.0   0.0   0.0   0.0
    

    I've PM you my IP address

  • TheLinuxBugTheLinuxBug Member
    edited February 2020

    @combahton_it said:

    @TheLinuxBug said:

    azbot said: I created a ticket to support 10 days ago. It was unbelievable that I was not supported yet, so I canceled the service.

    I will say this, I feel like their support quality and service quality has dropped off recently.

    What sort of support do you expect from a small instance which generates around 2,50€ of revenue per month? Sure, we might have been faster as we were smaller.

    All tickets are regulary handled within 24 hours, except accounting.

    A ticket I submitted got some canned reply blaming IO issues I reported on the product I purchased being an 'hdd' based product when IO to my server was <10M/sec and as far as I know, still is and hasn't been addressed. Instead they told me that I should upgrade to an SSD product instead of even reviewing my report. This is a very poor response to someone voicing an issue which likely also effects other customers on the same node.

    You dont have a server with SSD Storage, you just have a small HDD instance from a very old offer back a few years ago. Maybe it is swapping? Node shows not even 1% IOwait. Back then all instances had swap enabled.

    Below are pictures of our Grafana showing both the Nodes load and IOwait:


    I plan to discontinue the server after my current service period because of this and have already moved the sites I was hosting there away to another provider.

    Thats bad, because there are plans to introduce HDD Cloud Storage with NVMe Caching until the end of April, which means there will be no single nodes anymore. Higher redundancy, more I/O and customizable storage.

    Additionally, the bandwidth quality has also dropped off since I started using their services, however, I have taken this is stride so far as they have reportedly been upgrading their network.

    There were a few hickups from the end of December till mid of January. We got a unexpected big customer who increased the pressure on a dc interconnection, where we had to wait around four weeks for additional capacity. That capacity is now live: https://www.fastpipe.io/blog/?p=189

    Here is the problem with this. When I reported the issues initially there was an consistent IO wait on my server of greater than 30% while only moving data to my back-up server at less than 10M/sec because it couldn't get access to the data fast enough. When I ticketed originally, things were having issues. I understand you don't provide 24/7 support, however, I did expect such a report to be considered a little more important than "I will check on it tomorrow when I have time" and then also the response I got, instead of even considering my issue, was a blanket "You have HDD based server so this isn't unusual, you should upgrade" instead of someone actually even considering the issue, reviewing it, and responding with a comprehensive response like you finally did only after posting in this thread when it became apparent I wasn't happy.

    I shouldn't have to post here for my support requests to be taken at face value. The overall latency of the network and throughput of transfers has dropped off in general recently, so this already had me concerned before this IO issue appeared. The combination of those two issues and then the lazy response to my ticket request is what aggravated me enough to come here and post.

    Additionally, it isn't like I just signed-up for your services either, I believe I have been with you guys over a year now, I think after a year of being a customer and suggesting your services to others (as well as two of the sites in my signature on this site used to be hosted on that server), I have paid enough for you to take the time to actually take seriously / review what would be service impacting issues that are reported.

    I mean, until these issues started cropping up, you guys were one of the ideal providers... however, again... the way this was handled... the lackluster responses and accusation that I am not grateful for what I get, at the price I get it at, surely doesn't make me want to stay a customer. It will be a hard decision at renewal time. My recent anecdotal experience aside, the server had been stellar.

    my 2 cents.

    Cheers!

  • jh_aurologicjh_aurologic Member, Patron Provider

    @TheLinuxBug said:

    Additionally, the bandwidth quality has also dropped off since I started using their services, however, I have taken this is stride so far as they have reportedly been upgrading their network.

    That was related mostly to a dying router: https://status.combahton.net/incident/177

    Equipment has been replaced yesterday and there are already two ne QFX5100-48S-6Q on their way to rebuild redundancy.

  • The problem is still not resolved. Ive replied many times. Ive waited for so long. Just a waste of time of this continue

  • jh_aurologicjh_aurologic Member, Patron Provider

    @arjay316 said:
    The problem is still not resolved. Ive replied many times. Ive waited for so long. Just a waste of time of this continue

    The ticket is open since Saturday, means actually three working days. I will inform the responsible staff member and hope for a bit of patience.

  • edited February 2020

    Anyone reading this may consider my experience to be simply a one-off, and not at all representative of the service at Fastpipe.

    I have tried both SSD and HDD VMs. The HDD VM was very very slow, as others have observed. The SSD VM is not better either. Somehow, running apt-get install rsync emacs-nox on a freshly installed VM would result in the VM simply coming to a crawl (or maybe even stall, because nothing seems to be going). The operation (ie: apt-get install rsync emacs-nox) not completing in 30 minutes. Sure, there are some other packages (dependencies, updates) that apt-get is in turn installing:

    root@XXX:~# apt-get install rsync emacs-nox
    Reading package lists... Done
    Building dependency tree       
    Reading state information... Done
    The following additional packages will be installed:
      emacs-bin-common emacs-common emacs-el exim4-base exim4-config exim4-daemon-light guile-2.2-libs install-info libasound2 libasound2-data
      libevent-2.1-6 libfribidi0 libgc1c2 libgnutls-dane0 libgsasl7 libicu63 libjpeg62-turbo libkyotocabinet16v5 liblcms2-2 libltdl7 liblzo2-2
      libmailutils5 libmariadb3 libntlm0 libpython2.7 libpython2.7-minimal libpython2.7-stdlib libunbound8 libxml2 mailutils mailutils-common
      mariadb-common mysql-common psmisc
    Suggested packages:
      emacs-common-non-dfsg exim4-doc-html | exim4-doc-info eximon4 spf-tools-perl swaks libasound2-plugins alsa-utils dns-root-data liblcms2-utils
      mailutils-mh mailutils-doc
    The following NEW packages will be installed:
      emacs-bin-common emacs-common emacs-el emacs-nox exim4-base exim4-config exim4-daemon-light guile-2.2-libs install-info libasound2 libasound2-data
      libevent-2.1-6 libfribidi0 libgc1c2 libgnutls-dane0 libgsasl7 libicu63 libjpeg62-turbo libkyotocabinet16v5 liblcms2-2 libltdl7 liblzo2-2
      libmailutils5 libmariadb3 libntlm0 libpython2.7 libpython2.7-minimal libpython2.7-stdlib libunbound8 libxml2 mailutils mailutils-common
      mariadb-common mysql-common psmisc rsync
    0 upgraded, 36 newly installed, 0 to remove and 0 not upgraded.
    Need to get 58.0 MB of archives.
    After this operation, 242 MB of additional disk space will be used.
    Do you want to continue? [Y/n] 
    

    but it is still not expected that it would be so slow.

    On the dashboard, it shows "Memory 93.01% (1.86 GiB of 2 GiB)". This is not expected.

    Also, leaving aside the I/O problems, based on this test with nench.sh (yes, I am aware of the downsides/problems with such tests), it would suggest that the CPU is rather limited also:

    root@XXX:~# (curl -s wget.racing/nench.sh | bash; curl -s wget.racing/nench.sh | bash) 2>&1 | tee nench.log
    -------------------------------------------------
     nench.sh v2019.07.20 -- https://git.io/nench.sh
     benchmark timestamp:    2020-02-12 11:21:51 UTC
    -------------------------------------------------
    
    Processor:    Common KVM processor
    CPU cores:    1
    Frequency:    2499.998 MHz
    RAM:          1.9Gi
    Swap:         2.0Gi
    Kernel:       Linux 4.19.0-6-amd64 x86_64
    
    Disks:
    vda     30G  HDD
    
    CPU: SHA256-hashing 500 MB
        5.262 seconds
    CPU: bzip2-compressing 500 MB
        8.496 seconds
    CPU: AES-encrypting 500 MB
        3.391 seconds
    

    I have not raised a ticket with support. Combahton/Fastpipe charges monthly. So this is probably a VM to let expire, not to bother at all.

  • jh_aurologicjh_aurologic Member, Patron Provider

    @brightsunshine said:
    I have not raised a ticket with support. Combahton/Fastpipe charges monthly. So this is probably a VM to let expire, not to bother at all.

    Sure, thats a entry instance which has limited capabilities. If you need more power, you need to upgrade to a higher package.

  • @combahton_it said:
    Sure, thats a entry instance which has limited capabilities. If you need more power, you need to upgrade to a higher package.

    You're on LET, you know. Most do not upgrade and thus pay a higher price but instead look for another provider that offers more for less.

  • jh_aurologicjh_aurologic Member, Patron Provider

    @webcraft said:

    @combahton_it said:
    Sure, thats a entry instance which has limited capabilities. If you need more power, you need to upgrade to a higher package.

    You're on LET, you know. Most do not upgrade and thus pay a higher price but instead look for another provider that offers more for less.

    Sure thing! - I understand that. However, I'd say we do offer a lot for the budget pricing - I dont know any other company who offers a mix of all of the features we provide at that price tag.

  • edited February 2020

    @combahton_it Would you be able to at least offer something similar to the 2GB RAM VM from Hetzner or Netcup? Their offering is in the same price range, slightly cheaper, though not by much, but provides CPU and I/O that would not be as severely limited.

  • jh_aurologicjh_aurologic Member, Patron Provider

    There are plans to offer a new product line based on AMD CPUs within the next months. Also flexible Up/Downgrades and Storage (HDD, SSD, NVMe) is in our pipeline.

  • how about answer my question?

  • jh_aurologicjh_aurologic Member, Patron Provider

    @sibaper said:
    how about answer my question?

    Which?

  • Is Windows Server 2012 available? @combahton_it

  • jh_aurologicjh_aurologic Member, Patron Provider

    What is the exact question? Your comment just includes a statement, not a single question mark. If you mean the reason our looking glass was by the time of writing not showing the outbound route, I guess, the ddos-filters where active, which blocks icmp.

    Thanked by 1webcraft
Sign In or Register to comment.