Howdy, Stranger!

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


New SoYouStart 2018 Prices - Page 16
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.

New SoYouStart 2018 Prices

1131416181928

Comments

  • My FR arm box to my Netcup box:

    ------------------------------------------------------------
    [ ID] Interval       Transfer     Bandwidth
    [  3]  0.0- 1.0 sec  73.6 MBytes   618 Mbits/sec
    [  3]  1.0- 2.0 sec  64.4 MBytes   540 Mbits/sec
    [  3]  2.0- 3.0 sec  60.5 MBytes   508 Mbits/sec
    [  3]  3.0- 4.0 sec  51.2 MBytes   430 Mbits/sec
    [  3]  4.0- 5.0 sec  56.1 MBytes   471 Mbits/sec
    [  3]  5.0- 6.0 sec  60.9 MBytes   511 Mbits/sec
    [  3]  6.0- 7.0 sec  63.6 MBytes   534 Mbits/sec
    [  3]  7.0- 8.0 sec  57.9 MBytes   485 Mbits/sec
    [  3]  8.0- 9.0 sec  55.4 MBytes   465 Mbits/sec
    [  3]  9.0-10.0 sec  59.9 MBytes   502 Mbits/sec
    [  3]  0.0-10.0 sec   604 MBytes   506 Mbits/sec
    
    
    Thanked by 1dgprasetya
  • YorkiSYorkiS Member
    edited June 2018

    Guys, I won't able to let to go one in France DC.

    This is from GRA1 DC.

    # iperf -c ping.online.net -i 1
    ------------------------------------------------------------
    Client connecting to ping.online.net, TCP port 5001
    TCP window size: 43.8 KByte (default)
    ------------------------------------------------------------
    [  3] local x.x.x.x port 52750 connected with 62.210.18.40 port 5001
    [ ID] Interval       Transfer     Bandwidth
    [  3]  0.0- 1.0 sec  51.1 MBytes   429 Mbits/sec
    [  3]  1.0- 2.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  2.0- 3.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  3.0- 4.0 sec  37.4 MBytes   314 Mbits/sec
    [  3]  4.0- 5.0 sec  36.6 MBytes   307 Mbits/sec
    [  3]  5.0- 6.0 sec  36.5 MBytes   306 Mbits/sec
    [  3]  6.0- 7.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  7.0- 8.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  8.0- 9.0 sec  37.4 MBytes   314 Mbits/sec
    [  3]  9.0-10.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  0.0-10.0 sec   385 MBytes   323 Mbits/sec
    
    # iperf -c ping.online.net -i 1
    ------------------------------------------------------------
    Client connecting to ping.online.net, TCP port 5001
    TCP window size: 43.8 KByte (default)
    ------------------------------------------------------------
    [  3] local x.x.x.x port 52752 connected with 62.210.18.40 port 5001
    [ ID] Interval       Transfer     Bandwidth
    [  3]  0.0- 1.0 sec  69.0 MBytes   579 Mbits/sec
    [  3]  1.0- 2.0 sec  93.9 MBytes   787 Mbits/sec
    [  3]  2.0- 3.0 sec   122 MBytes  1.02 Gbits/sec
    [  3]  3.0- 4.0 sec   147 MBytes  1.23 Gbits/sec
    [  3]  4.0- 5.0 sec   172 MBytes  1.45 Gbits/sec
    [  3]  5.0- 6.0 sec   113 MBytes   951 Mbits/sec
    [  3]  6.0- 7.0 sec  37.4 MBytes   314 Mbits/sec
    [  3]  7.0- 8.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  8.0- 9.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  9.0-10.0 sec  37.2 MBytes   312 Mbits/sec
    [  3]  0.0-10.0 sec   867 MBytes   725 Mbits/sec
    
  • wonder how much money OVH loses with these boxes (when actually using the 250mbit)

  • YorkiS said: Guys, I won't able to let to go one in France DC.

    The same here. lol

  • MikeAMikeA Member, Patron Provider
    edited June 2018

    @mosan7763 said:
    wonder how much money OVH loses with these boxes (when actually using the 250mbit)

    I would guess other than media storage, quite a few would use them just to backup files from other OVH servers. If I didn't already have servers to backup to I would probably grab the cheapest one. OVH is too big to care about a server using 250Mbps anyways.

    Thanked by 1mtsbatalha
  • The U-Boot and its configuration are on a small 2MB flash chip on the server board. You can read this to get more knowledge of the boot process.

    It would be very unwise to change anything as this might also break the rescue boot and reinstall process.

    Thanked by 1vimalware
  • TheLinuxBugTheLinuxBug Member
    edited June 2018

    EricCartman said: The U-Boot and its configuration are on a small 2MB flash chip on the server board. You can read this to get more knowledge of the boot process.

    It would be very unwise to change anything as this might also break the rescue boot and reinstall process.

    Actually, what was more interesting to me and grabbed my attention was the fact that the installer places a 1000kb first partition on the drive which fdisk reports to be 'BIOS boot' ( http://prntscr.com/jxczsh ). I also wonder if this has something to do with options passed to the kernel however wasn't able to mount it yet. It may also be another 'uboot' at the beginning of the drive in case SPI fails. Not sure. Would be cool to find out but not sure I am willing to invest that much time into it.

    Cheers!

  • mkshmksh Member

    @TheLinuxBug said:
    I may bug a few of my buddies who work with ARM devices and kernels (for a living) later and see if they have any thoughts about it, but I am not holding my breathe as it doesn't seem OVH wishes to disclose that information or allow us to build kernels -- My request for kernel source got sent to the OS team and after that I have never heard back -- doubt I ever will.

    Inform the FSF? It's not up to OVH's goodwill to provide the sources.

    Thanked by 1Falzo
  • HarambeHarambe Member, Host Rep

    Well this could solve some of my temporary media server storage issues... Wonder how well the OpenMediaVault NFS works, would want to mount it directly in Proxmox.

    Thanked by 1vimalware
  • @Falzo finally got a reply from OVH:

    The source used to build ARM kernel 4.9.58-armada375 is available here: ftp://ftp.ovh.net/made-in-ovh/bzImage/linux-4.9.58_ARMv7.tgz

    There are no instructions unfortunately.

    Don't have time to mess with it atm but give it a try if you have time and let us know the outcome ;)

    Cheers!

  • Excellent find. /me looking at the /patches directory.

  • huckenhucken Member

    @sin said:
    What happened with the cheaper SYS servers that were supposed to be released?

    they keep adding and modifying the page for a week now. the old cheap ones got a disk or ram slapped on and got more expensive. so far no cheaper servers in sight only more expensive ones were added. we'll have to wait until they are finished.

  • mkshmksh Member
    edited June 2018

    @TheLinuxBug said:
    @Falzo finally got a reply from OVH:

    The source used to build ARM kernel 4.9.58-armada375 is available here: ftp://ftp.ovh.net/made-in-ovh/bzImage/linux-4.9.58_ARMv7.tgz

    There are no instructions unfortunately.

    Don't have time to mess with it atm but give it a try if you have time and let us know the outcome ;)

    Cheers!

    Without instructions or an obvious way to build it's still not really GPL compliant. Users must be able to produce a similar binary...

  • TheLinuxBugTheLinuxBug Member
    edited June 2018

    mksh said: Without instructions or an obvious way to build it's still not really GPL compliant. Users must be able to produce a similar binary...

    I am pretty sure this would be OVH's response:

    If they cared, in my ticket where I specifically asked for instructions they would have provided some, but as you can see, they them selves said, "There are no instructions unfortunately," so I wouldn't hold your breath. For that matter, they would have also participated in this thread, if they cared.

    I am actually surprised they replied to the ticket in the first place.

    Cheers!

  • harry_coxharry_cox Member
    edited June 2018

    On the SYS ARM server, using Ubuntu, did the fix to remove the armada kernel. Every time I use apt I get:
    Setting up initramfs-tools (0.122ubuntu8) ... update-initramfs: deferring update (trigger activated) Processing triggers for initramfs-tools (0.122ubuntu8) ... update-initramfs: Generating /boot/initrd.img-4.4.0-21-generic Unsupported platform. run-parts: /etc/initramfs/post-update.d//flash-kernel exited with return code 1 dpkg: error processing package initramfs-tools (--configure): subprocess installed post-installation script returned error exit status 1 Errors were encountered while processing: initramfs-tools E: Sub-process /usr/bin/dpkg returned an error code (1)

    Is there any way to fix this? Can I safely uninstall initramfs-tools?

  • mkshmksh Member

    @TheLinuxBug said:

    mksh said: Without instructions or an obvious way to build it's still not really GPL compliant. Users must be able to produce a similar binary...

    I am pretty sure this would be OVH's response:

    If they cared, in my ticket where I specifically asked for instructions they would have provided some, but as you can see, they them selves said, "There are no instructions unfortunately," so I wouldn't hold your breath. For that matter, they would have also participated in this thread, if they cared.

    I am actually surprised they replied to the ticket in the first place.

    Well, like i said it's not like they can really avoid it but i see how getting this through the thick skull of some support drone might be hard. Still if they didn't provide sources the FSF wouldn't find this the least bit of funny and OVH is a juicy target to make an example of. Same goes for the lack of instructions. If the that renders whatever they provide virtually non buildable it's sad for them as it's their problem to fix that. Non compliance basically revokes OVHs right to use or distribute the kernel. If they don't like that they shouldn't work with GPL'd code.

  • FalzoFalzo Member

    @harry_cox said:
    On the SYS ARM server, using Ubuntu, did the fix to remove the armada kernel. Every time I use apt I get:

    Errors were encountered while processing:
    initramfs-tools

    Is there any way to fix this? Can I safely uninstall initramfs-tools?

    as it does not use the 4.4 kernel at all, there is also no need for an initial ram disk. haven't tried, but pretty sure you can remove that alltogether.

  • FalzoFalzo Member

    @TheLinuxBug said:
    @Falzo finally got a reply from OVH:

    The source used to build ARM kernel 4.9.58-armada375 is available here: ftp://ftp.ovh.net/made-in-ovh/bzImage/linux-4.9.58_ARMv7.tgz

    There are no instructions unfortunately.

    Don't have time to mess with it atm but give it a try if you have time and let us know the outcome ;)

    Cheers!

    I had a first glance into the package. from what I can tell the .config is already in there and there is also a quite interesting folder inside, containing some patches. essentially that's providing some info about their 'development board' and patching stuff for the network driver and so on.

    also the tcp.c/.h tinkering mentioned earlier in this thread is in there. yet I did not find something obvious regarding the per connection speed limit. maybe it's somehow hidden like just providing some kind of trigger for QoS on the switch or the like, otherwise there should have been some algo to check on OVH internal/external network.
    maybe it even is some kind of weird bug related to their adjustments (which still would not be a reason why only external connections should be affected).

    if I have some time to spare during the weekend I am going to try and simply recompile their kernel first to see if that would work as a replacement and what's that doing about the speed limit.
    if I see some kind of success I might start digging deeper into it, trying to apply some of the patches to the official kernel (or even a later version) to get an own version to work with and eventually get rid of the speed limit issue.

  • @harry_cox said:
    On the SYS ARM server, using Ubuntu, did the fix to remove the armada kernel. Every time I use apt I get:
    Setting up initramfs-tools (0.122ubuntu8) ... update-initramfs: deferring update (trigger activated) Processing triggers for initramfs-tools (0.122ubuntu8) ... update-initramfs: Generating /boot/initrd.img-4.4.0-21-generic Unsupported platform. run-parts: /etc/initramfs/post-update.d//flash-kernel exited with return code 1 dpkg: error processing package initramfs-tools (--configure): subprocess installed post-installation script returned error exit status 1 Errors were encountered while processing: initramfs-tools E: Sub-process /usr/bin/dpkg returned an error code (1)

    Is there any way to fix this? Can I safely uninstall initramfs-tools?

    I tried this too, if you read the error, it's more serious. The 4.4.0 kernel you are trying to use isn't right. I undid mine and stuck with the 4.9 kernel. Funnily enough I was coming back on to see if anyone had managed to get it work with Ubuntu. :(

  • FalzoFalzo Member

    @michaels said:

    @harry_cox said:

    Is there any way to fix this? Can I safely uninstall initramfs-tools?

    I tried this too, if you read the error, it's more serious. The 4.4.0 kernel you are trying to use isn't right. I undid mine and stuck with the 4.9 kernel. Funnily enough I was coming back on to see if anyone had managed to get it work with Ubuntu. :(

    did you try to install ubuntu just without ticking the 'use distro kernel' box? at least that works for debian in my case without any further action needed.

    as we earlier agreed on the initial idea of installing another kernel from the repos is wrong, as it simply doesn't change anything. there is no grub but u-boot as loader, available kernels needs to be in a specific format which is not provided by the regular packages, so they are simply ignored.
    regardless if you install any kernel or not that way, it won't matter nor get used. if you remove the 4.9.58 kernel alltogether it should just net-boot an older version like 4.5.2 , which obviously doesn't have that speed limit issues.

    you don't need any other kernel packages from the repo, but instead of purging anything armada related like suggested earlier, you might want to keep the 4.5.2 modules (or reinstall them afterwards)...

    I know this thread got really messy and I apologize for it, please try to read again what was written first and how that evolved later on :-)

  • Shot2Shot2 Member
    edited June 2018

    However, it is worth noting this old kernel has no speed limit, but many other issues instead (including crypto-related trouble, e.g. kernel panics when using luks)

  • Falzo said: did you try to install ubuntu just without ticking the 'use distro kernel' box? at least that works for debian in my case without any further action needed.

    Yes, unfortunately (and I can't remember what) the OVH custom kernel broke something else. It's ok I am only using the box for nextcloud. Would just be nice to open it up! I might do a reinstall and switch to debian

  • TheLinuxBugTheLinuxBug Member
    edited June 2018

    Shot2 said: However, it is worth noting this old kernel has no speed limit, but many other issues instead (including crypto-related trouble, e.g. kernel panics when using luks)

    Actually, this is only the case if you setup LUKS on a newer kernel then go backwards. It seems the newer version isn't compatible with the older modules. If you re-initialize (luksFormat) the volume again on the 4.5.2 kernel after modprobing the modules you built, it will work normally and is stable.

    I had tried to go from 4.9.2 -> 4.5.2 and it didn't like that much.

    Cheers!

  • Shot2Shot2 Member

    Nope, I used LUKS on a running 4.5.2, but upon btrfs formatting a luks-encrypted device, kernel panic.

  • TheLinuxBugTheLinuxBug Member
    edited June 2018

    Shot2 said: Nope, I used LUKS on a running 4.5.2, but upon btrfs formatting a luks-encrypted device, kernel panic.

    Interesting, with ext4 I had no issues at all. Are you sure you have the module for brtfs loaded when doing so? Does it error with just formatting the volume w/ brtfs alone?

    A lot of the issues I ran into were caused by lack of modules and cryptsetup not being verbose enough about what it needed.

    If you need help compiling the modules, let me know and I will build whatever you need and package up a tar.gz of the source for you and send you a link.

    Cheers!

  • Shot2Shot2 Member
    edited June 2018

    No issue with btrfs, no issue with cryptsetup, only btrfs-over-cryptsetup fails.

    edit: actually, even ext4-over-cryptsetup fails.

    My modules loaded:
    # lsmod Module Size Used by sha256_arm 9692 0 cbc 2094 0 aes_arm 6500 0 hmac 2447 2 drbg 12842 1 ansi_cprng 3702 0 dm_crypt 16585 0 dm_mod 81421 1 dm_crypt ip6table_filter 943 0 ip6_tables 11345 1 ip6table_filter iptable_filter 1004 0 ip_tables 11090 1 iptable_filter x_tables 10502 4 ip6table_filter,ip_tables,iptable_filter,ip6_tables

    lsb_release -a No LSB modules are available. Distributor ID: Debian Description: Debian GNU/Linux 9.4 (stretch) Release: 9 Codename: stretch

    Linux hostname 4.5.2-armada375 #1 SMP Tue Oct 25 11:52:56 CEST 2016 armv7l GNU/Linux

  • twaintwain Member

    @michaels said:

    @harry_cox said:
    On the SYS ARM server, using Ubuntu, did the fix to remove the armada kernel. Every time I use apt I get:
    Setting up initramfs-tools (0.122ubuntu8) ... update-initramfs: deferring update (trigger activated) Processing triggers for initramfs-tools (0.122ubuntu8) ... update-initramfs: Generating /boot/initrd.img-4.4.0-21-generic Unsupported platform. run-parts: /etc/initramfs/post-update.d//flash-kernel exited with return code 1 dpkg: error processing package initramfs-tools (--configure): subprocess installed post-installation script returned error exit status 1 Errors were encountered while processing: initramfs-tools E: Sub-process /usr/bin/dpkg returned an error code (1)

    Is there any way to fix this? Can I safely uninstall initramfs-tools?

    I tried this too, if you read the error, it's more serious. The 4.4.0 kernel you are trying to use isn't right. I undid mine and stuck with the 4.9 kernel. Funnily enough I was coming back on to see if anyone had managed to get it work with Ubuntu. :(

    I am on Ubuntu 1604 using the template (distribution kernel ticked).

    I used @TheLinuxBug's suggested method of just removing /boot/$4.9.58-kernelfile, and then rebooted to get the netbooted 4.5.2 kernel.

    Didn't do anything else other than this; no removing or purging kernels, no installing any modules, nor symlinking anything. Apt working fine and haven’t seen any issues. Iptables also working fine and running fail2ban which is working fine as well.

    Thanked by 1Falzo
  • is a sale with no sale.how are we going to buy the servers on sale if there's no stock?

  • huckenhucken Member

    @ericnyamu said:
    is a sale with no sale.how are we going to buy the servers on sale if there's no stock?

    yeah, they are advertising the new servers for a week now on their page and for 2-3 weeks on twitter and still nothing but some price drops on game/storage

Sign In or Register to comment.