New ISO list is almost done syncing. We're going to try to update ISOs monthly for those marked as being latest version but mostly focusing on Ubuntu, Alma/Rocky, Debian.
Netboot
Ubuntu Desktop (22.04.1 LTS)
Ubuntu Server (22.04.1 LTS)
Debian (11.6)
Rocky (9.1)
Alma (9.1)
Stream (8)
Kali Linux (2202.4)
Kali Linux Live (2202.4)
FreeBSD (13.1)
Fedora Server (37)
Fedora CoreOS (37.20230122.3.0)
OpenBSD (7.2)
Alpine (3.17.2)
TinyCore
RancherOS
TrueNAS
Then these tools:
CloneZilla
GParted
SystemRescueCD
BootRepairDisk
Windows Offline NT Password & Registry Editor
They're all latest other than CentOS Stream, I'm not adding the 8.7GB version 9 unless anyone can find a minimal version, if that even exists. Then we're adding older versions, we'll do it four major versions back for all operating systems, the only ones I can't "guarantee" being all added are Windows.
CentOS 4.7, 5.9, 6.9, 7.9
Debian 7, 8, 9, 10
Ubuntu 14, 16, 18, 20
Windows 2012, 2016, 2019, 2022
They'll be maintained in waves, and either Alma or Rocky will replace the CentOS spot. So that most likely means we'll be missing Alma/Rocky 8 until the next "wave."
Wave 1
Ubuntu 14, Debian 7, CentOS 4, Windows 2012
Wave 2
Ubuntu 16, Debian 8, CentOS 5, Windows 2016
Wave 3
Ubuntu 18, Debian 9, CentOS 6, Windows 2019
Wave 4
Ubuntu 20, Debian 10, CentOS 7, Windows 2022
Its goes chronologically so the above four will be added in that order, then "Wave 5" will most likely be once the "Latest" at the top get mostly replaced maybe once a year or two. In the past we mostly kept ISO so people could download older versions but this means we're generally shifting towards keeping the latest versions and really old ones, but missing the "in between" stuff to some degree and those will end up mostly as templates.
Eventually we'll probably eliminate older "waves" by getting rid of every other one, so that could mean either Wave 1/2 disappears, then either Wave2/3, and so on but we might also selectively remove them per OS instead based on usage data.
We're probably going to mostly phase out custom ISO requests if it can be done through Netboot but I haven't really made any solid plans for that yet. Hopefully we'll have some automated system by then.
The couple of VMs I had on DALZ008, which have been down for a while, have been migrated to DALZ003 and DALZ010 and are available via VNC. They still have the old IPs, for now, and are not reachable via the interwebs yet, but I am glad to see the progress.
When these last two VMs are fully working then all 26 of my VirMach VMs will be online.
@FrankZ said:
The couple of VMs I had on DALZ008, which have been down for a while, have been migrated to DALZ003 and DALZ010 and are available via VNC. They still have the old IPs, for now, and are not reachable via the interwebs yet, but I am glad to see the progress.
When these last two VMs are fully working then all 26 of my VirMach VMs will be online.
New IPs assigned and the last two VMs are up!
All 26 of 26 VirMach VMs are now working as they should.
big day, a lot of news. First let's start with positive, via e-mail:
Dear VirMach Customers,
If you're receiving this email and still had an active service on node LAXA014 or TPAZ005, it has been re-created. Please check your service details page and "re-install" the operating system as it will most likely be required. If installation does not work, make sure to "power down" the VPS first and then attempt a re-install again but select a more recent operating system. If you notice any discrepancies or issues, please contact us and mention this email as well as any relavent information.
You have also been credited for the long duration outage. You can view your credits on the client area homepage, on the left sidebar under "Available Credit" section. This is in addition to any service extensions we may have applied previously.
Some important notes:
If you placed a cancellation request and it went through already, you may not receive credits or redeployment. In these cases feel free to contact us. If you have a cancellation request that has not yet completed, be sure to remove it if you wish to continue the service. You may need to do this quickly if the cancellation is due to occur. If the cancellation goes through, you can still always use the credits provided to make a new purchase.
If you require your data from the nodes, we are still working on recovery efforts but this may take a long time and it is not guaranteed to be available. To request a backup, please create a ticket with the title "Data Recovery Request" and provide the credentials to your new VPS. Let us know if you want us to dump the compressed backup inside your new VPS or if you prefer your new VPS being wiped/overwritten with the old data when/if available.
We apologize for the extreme inconvenience caused in these cases.
Thank you,
The VirMach Team
and now let's go to negative, via network status.
Emergency Maintenace https://billing.virmach.com/networkissues.php?view=nid1472
Thu, 23 Feb 2023 00:00:00 -0800
We're going to begin rolling reboots on certain nodes to avoid future issues. You may experience a few minute of downtime. We do not have an extensive list of servers affected or a timeline, as each server is being checked and may receive patches requiring a reboot.
Yeah, random reboots
Personal sad news:
I've got like 9$ in credits which is a total shock to me.... because my TPAZ005 instance was 2.5$ per year. My recreated instance is now 25$ per year and 9$ credits kinda make sense for few months of downtime... Yeah, I guess I will cancel that. In long term I made money on it ( @VirMach you sure this is how you wanted to do it? ), but this price jump is unacceptable !!11`11oneroneoeneoNEoneoneon!11eoneone
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
I recharged $20 to my account, and I want to pay the account upgrade service fee of $5 per month. But I can't pay, which reminds me that my account cannot be traded.My payment has been successful, and the invoice shows that it has not been paid. My payment credit has increased.
@risturiz said:
Finally TPAZ005 come back from the ashes and DAL online... Waiting for NYCB041 to go all green flag :-)
Slight delay on NYCB041. It was planned for the same day but pushed forward to probably Sunday or Monday.
@taoqi said:
I recharged $20 to my account, and I want to pay the account upgrade service fee of $5 per month. But I can't pay, which reminds me that my account cannot be traded.My payment has been successful, and the invoice shows that it has not been paid. My payment credit has increased.
If I understand correctly, you're marked as ineligible for purchase but want to purchase the extra support for your account? You'd need to appeal account issue and state that you want to resolve it and want to purchase additional support to facilitate the process and it will be considered.
We first need to check if you're linked to any accounts that opened a payment dispute before we can accept additional purchases.
This seems to be a new problem -
DALZ004 doesn't like 'rescue mode' - You can 'enable' rescue mode, but the VM stays in "offline" state. Disabling rescue mode brings it back online.
Fortunately, booting from the system rescue iso accomplishes pretty much the same thing aside from SSH access
Small update on NYCB009 (shared hosting) for anyone who's interested in asking any questions, I wouldn't normally go over it on the network status page as it would just confuse people but we did mention it to some degree there.
We've contacted engineering team for our RAID controller and waiting for their response, but what's happening is that it's overloading after a disk drops off within the hardware RAID controller. We're locked to specific settings to make this function so I can't apply any normal "fixes" without it running into other issues. The NVMe itself seems fine and this seems to have popped up after a kernel update, meaning the drivers most likely need to be patched by the manufacturer. The normal behavior would be for the disk to drop off and then be able to add it back in/rescan and restore it, but while a rescan brings the disk back in view on the server, it doesn't appear in the controller. The other expected behavior would be that if one disk drops off the controller should be able to carry on in critical mode but instead it's completely overloading, and this is causing internal server error with WHM/cPanel.
So basically it's what I originally feared and why we decided against HW RAID a long time ago (plus there were other problems at the time such as supply chain issues, as in we couldn't even order more in time.) We do have backups set up and also actually have enough SATA SSDs to move everyone there and have everything served from those instead should the situation get worse. After firmware update and patching kernel as well as BIOS (IIRC) it did appear to be more stable for some time but obviously it ran into issues again recently.
@erk said:
This seems to be a new problem -
DALZ004 doesn't like 'rescue mode' - You can 'enable' rescue mode, but the VM stays in "offline" state. Disabling rescue mode brings it back online.
Fortunately, booting from the system rescue iso accomplishes pretty much the same thing aside from SSH access
There's two VMs offline here, I'm taking a quick look in case there's a major issue that could affect others.
I've verified the issue, I know the fix, doing it now.
One if my vms.. has a different ip on Whmcs vs SolusVM.
I think because of this something really bad is happening and I cannot seem to get the install happen...
@taoqi said:
I recharged $20 to my account, and I want to pay the account upgrade service fee of $5 per month. But I can't pay, which reminds me that my account cannot be traded.My payment has been successful, and the invoice shows that it has not been paid. My payment credit has increased.
If I understand correctly, you're marked as ineligible for purchase but want to purchase the extra support for your account? You'd need to appeal account issue and state that you want to resolve it and want to purchase additional support to facilitate the process and it will be considered.
We first need to check if you're linked to any accounts that opened a payment dispute before we can accept additional purchases.
Yes. I want to solve this problem. Please help me. How should I appeal my account?This is the payment slip number.This is the payment slip number #1537042
I had an issue from the flash sales where I got provisioned the deal after the one I had purchased and I just wanna say that Virmach actually re-provisioned to the specifications that I purchased at.
This was done in a decent timely manner, so well done Virmach.
@VirMach
Invoice #1413533
My VPS is a pre -sale of Tokyo for two years. Now my node has become AMSD030X.
It has been offline for two months. Please help me change the VPS back to Tokyo node. Thank you
No comment on Tokyo other than some decisions/changes need to be implemented and obviously we didn't end up running that sale. There won't be another Tokyo sale for now until we sort out existing Tokyo things for existing customers.
@imok said:
Issues with TPAZ002? My VPS keeps going offline.
Tampa pretty much has had issues sprinkled on everything, I've been managing it without actually making datacenter hands requests as I'm afraid it could be a repeat of every other hands request we've made in Tampa in the past.
However, I don't think that should be happening, let me know additional details. I mean the usage is suspiciously low but I figured that location has no use case other than idling as usage is low on pretty much all of it all the time. I do actually think almost everyone that went to Tampa just has like 30 virtual servers around the world and don't use it though. Load averages are 0.9, 0.6, and 8 right now, you can check the graphs.
Also I think that's the only reason TPAZ002 is alive right now. It'd be funny if you're the only 1 person out of 150~ that decided to actually use it and when you do it just crashes the VM.
Anyway pretty much all of Tampa got screwed up during shipping since one datacenter sent it out like 3 servers per box with what I assume to be no padding since I don't know how else it'd fit, and the receiving datacenter also wanted to send out 3 servers per box back to us. The second datacenter did heavy troubleshooting, plus there's an entire screenplay I wrote that's relevant.
These servers also had a lot of difficulties in the past, but of course my opinion on the matter is biased as they obviously got sent out healthy and then a lot of CPU re-seating happening. I believe there may have been plenty of rounds of CPU pin bending and unbending as well as CPU socket violation.
Then the motherboards may have taken a beating. I just got the rest of Tampa back this week, I'm going to work on them and the plan is to just do minimal poking to the existing three because next touching of server is essentially a pretty good guarantee of hardware destruction and several weeks delay.
TLDR: I recommend you just contact us to move it to Miami if it's causing grief. We were supposed to make a decision pertaining to the location around 2 months ago probably but then everything else happens so I recommend you make the decision yourself in this case.
@VirMach said: I recommend you just contact us to move it to Miami if it's causing grief.
Thanks, I will do it when I finish the Ubuntu release upgrade. Weird thing is it finished successfully before the server went down. Maybe another version just appeared minutes after finishing, because a rollback of the VM can't be possible... right? right?
I was rebuilding Discourse when the server went off.
EDIT: Everything finished successfully. Server still up. I will wait for the next downtime because I'm feeling too lazy right now and I don't care about my visitors.
@VirMach said:
New ISO list is almost done syncing. We're going to try to update ISOs monthly for those marked as being latest version but mostly focusing on Ubuntu, Alma/Rocky, Debian.
Netboot
Ubuntu Desktop (22.04.1 LTS)
Ubuntu Server (22.04.1 LTS)
Debian (11.6)
Rocky (9.1)
Alma (9.1)
Stream (8)
Kali Linux (2202.4)
Kali Linux Live (2202.4)
FreeBSD (13.1)
Fedora Server (37)
Fedora CoreOS (37.20230122.3.0)
OpenBSD (7.2)
Alpine (3.17.2)
TinyCore
RancherOS
TrueNAS
The Alpine ISO above is "Standard" instead of "Virtual", which is the preferred one really. They only differ in the former having 200MB or so of hardware drivers which are not used in a virtual environment. If you want to save memory you can use export DISKOPTS="-k virt" before you run setup-alpine to get the slimmed down kernel and avoid the (large) firmware package being installed.
Comments
Are you sure this time?
Surely that's not all nodes.
Free NAT KVM | Free NAT LXC
i think virmach found the solution to 5900X and Asrock Rack X570! Hope he applies it to all boxes with same config.
I bench YABS 24/7/365 unless it's a leap year.
ALL! 100% GREEN
I bench YABS 24/7/365 unless it's a leap year.
Thus, the end is nigh.
Whatever that means, the end is nigh.
Yeah.
♻ Amitz day is October 21.
♻ Join Nigh sect by adopting my avatar. Let us spread the joys of the end.
Well TPAZ005 is gone and DALZ008 is locked
Too much like OGF when there are four threads regarding the same provider on the front page. Merged.
TPAZ005 is 3950X? thats a rare one.
hard to tell now since the client area announcement is removed.
I bench YABS 24/7/365 unless it's a leap year.
The old fart shuts me down when I'm compiling QT6
Ontario Dildo Inspector
I have not found that page to reflect the reality of individual VMs being online whatsoever.
Dataplane.org's current server hosting provider list
New ISO list is almost done syncing. We're going to try to update ISOs monthly for those marked as being latest version but mostly focusing on Ubuntu, Alma/Rocky, Debian.
Then these tools:
They're all latest other than CentOS Stream, I'm not adding the 8.7GB version 9 unless anyone can find a minimal version, if that even exists. Then we're adding older versions, we'll do it four major versions back for all operating systems, the only ones I can't "guarantee" being all added are Windows.
They'll be maintained in waves, and either Alma or Rocky will replace the CentOS spot. So that most likely means we'll be missing Alma/Rocky 8 until the next "wave."
Wave 1
Ubuntu 14, Debian 7, CentOS 4, Windows 2012
Wave 2
Ubuntu 16, Debian 8, CentOS 5, Windows 2016
Wave 3
Ubuntu 18, Debian 9, CentOS 6, Windows 2019
Wave 4
Ubuntu 20, Debian 10, CentOS 7, Windows 2022
Its goes chronologically so the above four will be added in that order, then "Wave 5" will most likely be once the "Latest" at the top get mostly replaced maybe once a year or two. In the past we mostly kept ISO so people could download older versions but this means we're generally shifting towards keeping the latest versions and really old ones, but missing the "in between" stuff to some degree and those will end up mostly as templates.
Eventually we'll probably eliminate older "waves" by getting rid of every other one, so that could mean either Wave 1/2 disappears, then either Wave2/3, and so on but we might also selectively remove them per OS instead based on usage data.
We're probably going to mostly phase out custom ISO requests if it can be done through Netboot but I haven't really made any solid plans for that yet. Hopefully we'll have some automated system by then.
Damn, I'm riding 'restart waves'
Every couple minutes the server will restart causing high server load.
No email notification so far.
https://microlxc.net/
The couple of VMs I had on DALZ008, which have been down for a while, have been migrated to DALZ003 and DALZ010 and are available via VNC. They still have the old IPs, for now, and are not reachable via the interwebs yet, but I am glad to see the progress.
When these last two VMs are fully working then all 26 of my VirMach VMs will be online.
Finally provisioned after almost 2 months on DALZ004
If it does not live up to your expectations, feel free to shoot me a PM and I'll take it off your hands.
i was late
I bench YABS 24/7/365 unless it's a leap year.
Apologies, if I sounded not happy. I am good 😊
My usage and providers are these
12 VPS for learning? That's a lot 🤔
All 12 are not mine. Didn’t quote the original post but there were 12 VPS pending from FLASH sales since Jan
Pending is much more pro than idling.
the TYC040 node has been active for 3 days and has been locked again
wow, seems TYOC007 is accessible now.
New IPs assigned and the last two VMs are up!
All 26 of 26 VirMach VMs are now working as they should.
big day, a lot of news. First let's start with positive, via e-mail:
and now let's go to negative, via network status.
Yeah, random reboots
Personal sad news:
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
:'(
Finally TPAZ005 come back from the ashes and DAL online... Waiting for NYCB041 to go all green flag :-)
I recharged $20 to my account, and I want to pay the account upgrade service fee of $5 per month. But I can't pay, which reminds me that my account cannot be traded.My payment has been successful, and the invoice shows that it has not been paid. My payment credit has increased.
Slight delay on NYCB041. It was planned for the same day but pushed forward to probably Sunday or Monday.
If I understand correctly, you're marked as ineligible for purchase but want to purchase the extra support for your account? You'd need to appeal account issue and state that you want to resolve it and want to purchase additional support to facilitate the process and it will be considered.
We first need to check if you're linked to any accounts that opened a payment dispute before we can accept additional purchases.
This seems to be a new problem -
DALZ004 doesn't like 'rescue mode' - You can 'enable' rescue mode, but the VM stays in "offline" state. Disabling rescue mode brings it back online.
Fortunately, booting from the system rescue iso accomplishes pretty much the same thing aside from SSH access
i am happy i can see
LAXA029
back online
virvir High5 , Thank you.
all that is missing now is the migrate button and we are men again.
Small update on NYCB009 (shared hosting) for anyone who's interested in asking any questions, I wouldn't normally go over it on the network status page as it would just confuse people but we did mention it to some degree there.
We've contacted engineering team for our RAID controller and waiting for their response, but what's happening is that it's overloading after a disk drops off within the hardware RAID controller. We're locked to specific settings to make this function so I can't apply any normal "fixes" without it running into other issues. The NVMe itself seems fine and this seems to have popped up after a kernel update, meaning the drivers most likely need to be patched by the manufacturer. The normal behavior would be for the disk to drop off and then be able to add it back in/rescan and restore it, but while a rescan brings the disk back in view on the server, it doesn't appear in the controller. The other expected behavior would be that if one disk drops off the controller should be able to carry on in critical mode but instead it's completely overloading, and this is causing internal server error with WHM/cPanel.
So basically it's what I originally feared and why we decided against HW RAID a long time ago (plus there were other problems at the time such as supply chain issues, as in we couldn't even order more in time.) We do have backups set up and also actually have enough SATA SSDs to move everyone there and have everything served from those instead should the situation get worse. After firmware update and patching kernel as well as BIOS (IIRC) it did appear to be more stable for some time but obviously it ran into issues again recently.
There's two VMs offline here, I'm taking a quick look in case there's a major issue that could affect others.
I've verified the issue, I know the fix, doing it now.
(edit) Brought the two back online.
One if my vms.. has a different ip on Whmcs vs SolusVM.
I think because of this something really bad is happening and I cannot seem to get the install happen...
Sigh... one day...
Yes. I want to solve this problem. Please help me. How should I appeal my account?This is the payment slip number.This is the payment slip number #1537042
A Virmach Praise post:
I had an issue from the flash sales where I got provisioned the deal after the one I had purchased and I just wanna say that Virmach actually re-provisioned to the specifications that I purchased at.
This was done in a decent timely manner, so well done Virmach.
Well, please ignore my reply. I have solved the problem myself
@VirMach
Invoice #1413533
My VPS is a pre -sale of Tokyo for two years. Now my node has become AMSD030X.
It has been offline for two months. Please help me change the VPS back to Tokyo node. Thank you
All praise @VirMach!!!!!!
the transfer to @ariq01 has been completed, maybe I will see what other Flash Sale servers I have not yet used will be up for trade
this is 100% up for trading though.
Does anyone else remember this? GuGuGu~~
https://lowendspirit.com/discussion/comment/126562/#Comment_126562
Issues with TPAZ002? My VPS keeps going offline.
No comment on Tokyo other than some decisions/changes need to be implemented and obviously we didn't end up running that sale. There won't be another Tokyo sale for now until we sort out existing Tokyo things for existing customers.
Tampa pretty much has had issues sprinkled on everything, I've been managing it without actually making datacenter hands requests as I'm afraid it could be a repeat of every other hands request we've made in Tampa in the past.
However, I don't think that should be happening, let me know additional details. I mean the usage is suspiciously low but I figured that location has no use case other than idling as usage is low on pretty much all of it all the time. I do actually think almost everyone that went to Tampa just has like 30 virtual servers around the world and don't use it though. Load averages are 0.9, 0.6, and 8 right now, you can check the graphs.
Also I think that's the only reason TPAZ002 is alive right now. It'd be funny if you're the only 1 person out of 150~ that decided to actually use it and when you do it just crashes the VM.
Anyway pretty much all of Tampa got screwed up during shipping since one datacenter sent it out like 3 servers per box with what I assume to be no padding since I don't know how else it'd fit, and the receiving datacenter also wanted to send out 3 servers per box back to us. The second datacenter did heavy troubleshooting, plus there's an entire screenplay I wrote that's relevant.
These servers also had a lot of difficulties in the past, but of course my opinion on the matter is biased as they obviously got sent out healthy and then a lot of CPU re-seating happening. I believe there may have been plenty of rounds of CPU pin bending and unbending as well as CPU socket violation.
Then the motherboards may have taken a beating. I just got the rest of Tampa back this week, I'm going to work on them and the plan is to just do minimal poking to the existing three because next touching of server is essentially a pretty good guarantee of hardware destruction and several weeks delay.
TLDR: I recommend you just contact us to move it to Miami if it's causing grief. We were supposed to make a decision pertaining to the location around 2 months ago probably but then everything else happens so I recommend you make the decision yourself in this case.
Thanks, I will do it when I finish the Ubuntu release upgrade. Weird thing is it finished successfully before the server went down. Maybe another version just appeared minutes after finishing, because a rollback of the VM can't be possible... right? right?
I was rebuilding Discourse when the server went off.
EDIT: Everything finished successfully. Server still up. I will wait for the next downtime because I'm feeling too lazy right now and I don't care about my visitors.
(into ) Update 2/27/2022 - We are performing some of these changes in Frankfurt at this time.
NO TOUCHY MY FRANKFURT!!!1111oneoneone
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
@imok I moved my VM on TPAZ005 to Miami last August and have been happy I did so.
The Alpine ISO above is "Standard" instead of "Virtual", which is the preferred one really. They only differ in the former having 200MB or so of hardware drivers which are not used in a virtual environment. If you want to save memory you can use
export DISKOPTS="-k virt"before you runsetup-alpineto get the slimmed down kernel and avoid the (large) firmware package being installed.Frankfurt getting rebooted.
btw. DRAGON BALL XENOVERSE from your Steam wishlist is now on sale!
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
Just noticed with VPS at rebooted FFME007:
Forgot to disable it? :P