@FrankZ said: Average disk write over the last 48 hours was 106 MB/s.
Average disk read rate over the last 48 hours was 101 MB/s
You what mate?
Abuser! You sure those number makes sense? :P
Good catch, I am sure that folks on that node will be happier after I leave.
Actually the disk tests only runs every 6 hours, I will edit the OP to reflect that.
Been waiting for someone to post up an offer for that Tokyo storage node and somehow missed it!
Would like something closer to 2TB, but I'll honestly take what I can get. Imagine if the current storage offer was available in Tokyo... One can dream I guess.
@theloafingone said:
Been waiting for someone to post up an offer for that Tokyo storage node and somehow missed it!
Would like something closer to 2TB, but I'll honestly take what I can get. Imagine if the current storage offer was available in Tokyo... One can dream I guess.
@FrankZ was offering to transfer a Tokyo node just yesterday...
@theloafingone said:
Been waiting for someone to post up an offer for that Tokyo storage node and somehow missed it!
Would like something closer to 2TB, but I'll honestly take what I can get. Imagine if the current storage offer was available in Tokyo... One can dream I guess.
@FrankZ was offering to transfer a Tokyo node just yesterday...
Yes, I was referring to that with the "missed it" part. Next time, gadget!
@theloafingone said:
Been waiting for someone to post up an offer for that Tokyo storage node and somehow missed it!
Would like something closer to 2TB, but I'll honestly take what I can get. Imagine if the current storage offer was available in Tokyo... One can dream I guess.
@FrankZ was offering to transfer a Tokyo node just yesterday...
Yes, I was referring to that with the "missed it" part. Next time, gadget!
Oh, did he transfer it to US region already? Too bad. Good luck with your next hunt.
@rkzed said:
does anyone on node TYOC005 have same issue with me? the disk is missing (undetected?) from VPS.
If it shows something like "No boot device available" in VNC, do not reinstall the operating system. This thing has happened to me before. It will get fixed automatically by VirMach if you do not reinstall, otherwise IIRC it needs to be fixed manually which takes longer.
@somik said: Oh, did he transfer it to US region already? Too bad. Good luck with your next hunt.
It is in the process of being transferred to a helpful member who wanted Tokyo storage.
Hope you have better luck next time @theloafingone
@FrankZ said: If it shows something like "No boot device available" in VNC, do not reinstall the operating system. This thing has happened to me before. It will get fixed automatically by VirMach if you do not reinstall, otherwise IIRC it needs to be fixed manually which takes longer.
yes, exactly like you described. already put a ticket with them. hope it gets resolved soon. this is the first major problem with the VPS, otherwise, it's been stable for me.
It's all good. I'll make due with this banger of a biennial storage special in LA (2c/2gb/4TB@$2.67/m) for now since it has a decently good ping to Japan.
On a somewhat related note, my initial install seems to be broken and the instructions in the introduction e-mail are broken. I'm not an expert on VPS installations and the previous discussion about this was pretty sparse. Tried submitting a ticket with a screenshot but it looks like the final submit button is bugged - it keeps asking me to open (attach?) a file. VNC is showing the attached screen.
Any help would be much appreciated. Thanks for your time.
@theloafingone - I get a "the image can not be displayed because it contains errors" error when trying to view the image you posted so I can't really see what you are talking about.
@FrankZ said: @theloafingone - I get a "the image can not be displayed because it contains errors" error when trying to view the image you posted so I can't really see what you are talking about.
That's really strange, the error isn't showing on my side. Here's the direct imgur link: https://i.imgur.com/qiOQweK.png
But basically it says "Welcome to Grub! Error: no such partition" then the prompt is saying "grub rescue>"
@theloafingone - I expect it was delivered without an O/S installed. With a 4TB VM I would install from an ISO so as to make at least three partitions, one for boot, one for the O/S, and one for data. You can do this by going to the billing panel and selecting the VM, then select the "VPS Control Panel" button. This will take you to the SolusVM panel. Click "Manage" on the line with your VM (this should be the only VM available). On the next page ensure that the "Boot Order" shows "(1) CDROM (2) Hard Disk", if not, select this option. Then select "CDROM" from the line above that,. and below the buttons. This will take you to a new display on the bottom of the page where you can use the select box to select the ISO for the O/S you wish to install. After selected, click the mount button. Now click the "Reboot" button. After your VM restarts you should be able to enter into the VM via VNC and see the prompts for the ISO install.
Note: If you do not know the network information for this VM then select DHCP when setting up the network and the network information should be provided automatically.
After you have installed the O/S in the VM you should get a final prompt to restart the VM. At this time go back to the SolusVM panel and change the "Boot Order" to "Hard Drive Only". Go back to the "CDROM" and click the "Unmount" button to unmount the ISO that you used to install the O/S. Then click the "Reboot" button to restart the VM. If you do not use the "Reboot" button in SolusVM to restart the ISO will not unmount and the boot order will not change. So if you click the "Restart" button shown at the end of the ISO install you will still need to click the "ReBoot" button in SolusVM other wise you will just boot right back into the ISO install.
I'll stop here just in case I am on the wrong track.
EDIT: The image you posted shows just fine in Brave browser, but still gives me the error I described above in FireFox, strange.
@Flying_Chinaman said: Then ya can see that typical oversold CPU steal burst too,
Just wonder how many storage offers the old fart sold on that single node
We don't oversell the storage amounts at all so it ends up being under-utilized. Tokyo pretty much got the best RAID controller and drives available at the time.
I believe we actually undersold memory for some time, then we oversold that, then we adjusted it back down a few times. So these old plans are a little heavy on memory but we only did that because we basically could since for years the storage nodes had very little RAM usage and a lot of it. As for CPU, yeah it's obviously oversold.
The only plans I know of that actually look like they're not oversold on CPU would probably be BuyVM slices, and that's 4Gb RAM to 1 Core ratio. Anything higher than that pretty much from anyone on Ryzen is oversold, basic math. So ours looks about 3x "oversold" for the 4TB package if you go based off that and ignore actual usage levels. Anyway, let's quickly talk about the actual reason.
@FrankZ said: Checking every 2 minutes the ....
Average steal rate over the last 48 hours was 5.29%, bursting to a high of 29.30% .
Average wait state over the last 48 hours was 0.49%.
Average packet loss over the last 48 hours was 0.28%
Disk test run every 6 hours....
Average disk write over the last 48 hours was 106 MB/s.
Average disk read rate over the last 48 hours was 101 MB/s
Seems fine to me for a storage server.
I recently identified a pattern and have a good idea for a solution to one of the multiple problems on this node, I've already covered the others in the past but there's definitely something up with it. Once it's complete it'll improve it. Not sure by how much. This is in terms of CPU steal "bursts."
I don't have a timeline and that's why nothing's been sent out yet. It's going to probably end up being an expensive hands request and take some time to complete with possible side effects so it has to be arranged properly.
External monitoring shows my VPS went offline around 18 May 04:08 UTC. Can't access control panel due to the dreaded "Operation Timed Out After 90001 Milliseconds With 0 Bytes Received" error. https://status.virm.ac/ shows no recent data for CHIZ001 (or anything at CHI, actually), last updated 7 hours ago (which seems to sync with my external monitoring alert). Nothing listed on the network status page.
@JBB - You're not alone as it seems that all of Chicago DC #1 (CHIZ001, CHIZ002, CHIZ003, CHIZ004) has dropped. I'm on CHIZ003 and am getting the same "Operation Timed Out ..." error as you are. I expect it may be a DC network issue, or maybe someone unplugged the cable to VirMach's rack because you know VirMach does not have the best luck when it comes to DC hands.
@FrankZ said: @JBB - Your not alone as it seems that all of Chicago DC #1 (CHIZ001, CHIZ002, CHIZ003, CHIZ004) has dropped.
Thanks, I suspected as much.
It's been a reliable location for me so far, but I have noticed that under interactive use sometimes it feels like the network is a bit intermittent. I'll type commands in my term and nothing for a second or two, then it catches back up. I've never been able to measure it as packet loss, it's both more subtle and more annoying than that.
@FrankZ said: @JBB - Your not alone as it seems that all of Chicago DC #1 (CHIZ001, CHIZ002, CHIZ003, CHIZ004) has dropped. I'm on CHIZ003 and am getting the same "Operation Timed Out ..." error as you are. I expect it may be a DC network issue, or maybe someone unplugged the cable to VirMach's rack because you know VirMach does not have the best luck when it comes to DC hands.
I guess this means the VPS of mine that's vanished is in Chicago, then
(doesn't seem to be a way to see what node you're on once it's dead... /sigh)
@FrankZ said: @JBB - Your You're not alone as it seems that all of Chicago DC #1 (CHIZ001, CHIZ002, CHIZ003, CHIZ004) has dropped. I'm on CHIZ003 and am getting the same "Operation Timed Out ..." error as you are. I expect it may be a DC network issue, or maybe someone unplugged the cable to VirMach's rack because you know VirMach does not have the best luck when it comes to DC hands.
Comments
You what mate?
Abuser! You sure those number makes sense? :P
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
https://lowendspirit.com/uploads/editor/gi/ippw0lcmqowk.png
Good catch, I am sure that folks on that node will be happier after I leave.
Actually the disk tests only runs every 6 hours, I will edit the OP to reflect that.
LES • About • Donate • Rules • Support
My first shutdown "we permit you to average around 80 TPS/IOPS" xD ... So, i cant run netmaker test server on Virmach :-|
Not really clued in on Netmaker but why is a VPN service generating so much disk IO?
Don't know either ... Was a test system with 5 clients ... Maybe a bug? xD
Been waiting for someone to post up an offer for that Tokyo storage node and somehow missed it!
Would like something closer to 2TB, but I'll honestly take what I can get. Imagine if the current storage offer was available in Tokyo... One can dream I guess.
@FrankZ was offering to transfer a Tokyo node just yesterday...
Somik.org - Server admins cheat codes
Yes, I was referring to that with the "missed it" part. Next time, gadget!
Oh, did he transfer it to US region already? Too bad. Good luck with your next hunt.
Somik.org - Server admins cheat codes
does anyone on node TYOC005 have same issue with me? the disk is missing (undetected?) from VPS.
If it shows something like "No boot device available" in VNC, do not reinstall the operating system. This thing has happened to me before. It will get fixed automatically by VirMach if you do not reinstall, otherwise IIRC it needs to be fixed manually which takes longer.
It is in the process of being transferred to a helpful member who wanted Tokyo storage.
Hope you have better luck next time @theloafingone
LES • About • Donate • Rules • Support
yes, exactly like you described. already put a ticket with them. hope it gets resolved soon. this is the first major problem with the VPS, otherwise, it's been stable for me.
It's all good. I'll make due with this banger of a biennial storage special in LA (2c/2gb/4TB@$2.67/m) for now since it has a decently good ping to Japan.
On a somewhat related note, my initial install seems to be broken and the instructions in the introduction e-mail are broken. I'm not an expert on VPS installations and the previous discussion about this was pretty sparse. Tried submitting a ticket with a screenshot but it looks like the final submit button is bugged - it keeps asking me to open (attach?) a file. VNC is showing the attached screen.
Any help would be much appreciated. Thanks for your time.
@theloafingone - I get a "the image can not be displayed because it contains errors" error when trying to view the image you posted so I can't really see what you are talking about.
LES • About • Donate • Rules • Support
That's really strange, the error isn't showing on my side. Here's the direct imgur link: https://i.imgur.com/qiOQweK.png
But basically it says "Welcome to Grub! Error: no such partition" then the prompt is saying "grub rescue>"
@theloafingone - I expect it was delivered without an O/S installed. With a 4TB VM I would install from an ISO so as to make at least three partitions, one for boot, one for the O/S, and one for data. You can do this by going to the billing panel and selecting the VM, then select the "VPS Control Panel" button. This will take you to the SolusVM panel. Click "Manage" on the line with your VM (this should be the only VM available). On the next page ensure that the "Boot Order" shows "(1) CDROM (2) Hard Disk", if not, select this option. Then select "CDROM" from the line above that,. and below the buttons. This will take you to a new display on the bottom of the page where you can use the select box to select the ISO for the O/S you wish to install. After selected, click the mount button. Now click the "Reboot" button. After your VM restarts you should be able to enter into the VM via VNC and see the prompts for the ISO install.
Note: If you do not know the network information for this VM then select DHCP when setting up the network and the network information should be provided automatically.
After you have installed the O/S in the VM you should get a final prompt to restart the VM. At this time go back to the SolusVM panel and change the "Boot Order" to "Hard Drive Only". Go back to the "CDROM" and click the "Unmount" button to unmount the ISO that you used to install the O/S. Then click the "Reboot" button to restart the VM. If you do not use the "Reboot" button in SolusVM to restart the ISO will not unmount and the boot order will not change. So if you click the "Restart" button shown at the end of the ISO install you will still need to click the "ReBoot" button in SolusVM other wise you will just boot right back into the ISO install.
I'll stop here just in case I am on the wrong track.
EDIT: The image you posted shows just fine in Brave browser, but still gives me the error I described above in FireFox, strange.
LES • About • Donate • Rules • Support
I'm on firefox and it shows just fine for me... weird...
Somik.org - Server admins cheat codes
Good to know. I expect the problem is probably on my end.
LES • About • Donate • Rules • Support
How could the old fart automatically fix things
Any masochist would setup LUKS + LVM on your VPS?
Yo, join our premium masochist club
I think you need to tighten back your screws...
Somik.org - Server admins cheat codes
That will be the day.
@Flying_Chinaman - It is VirBot's day job to fix things like this.
LES • About • Donate • Rules • Support
We don't oversell the storage amounts at all so it ends up being under-utilized. Tokyo pretty much got the best RAID controller and drives available at the time.
I believe we actually undersold memory for some time, then we oversold that, then we adjusted it back down a few times. So these old plans are a little heavy on memory but we only did that because we basically could since for years the storage nodes had very little RAM usage and a lot of it. As for CPU, yeah it's obviously oversold.
The only plans I know of that actually look like they're not oversold on CPU would probably be BuyVM slices, and that's 4Gb RAM to 1 Core ratio. Anything higher than that pretty much from anyone on Ryzen is oversold, basic math. So ours looks about 3x "oversold" for the 4TB package if you go based off that and ignore actual usage levels. Anyway, let's quickly talk about the actual reason.
I recently identified a pattern and have a good idea for a solution to one of the multiple problems on this node, I've already covered the others in the past but there's definitely something up with it. Once it's complete it'll improve it. Not sure by how much. This is in terms of CPU steal "bursts."
I don't have a timeline and that's why nothing's been sent out yet. It's going to probably end up being an expensive hands request and take some time to complete with possible side effects so it has to be arranged properly.
Anyone else on CHIZ001 down?
External monitoring shows my VPS went offline around 18 May 04:08 UTC. Can't access control panel due to the dreaded "Operation Timed Out After 90001 Milliseconds With 0 Bytes Received" error. https://status.virm.ac/ shows no recent data for CHIZ001 (or anything at CHI, actually), last updated 7 hours ago (which seems to sync with my external monitoring alert). Nothing listed on the network status page.
Looking glass at https://chi.lg.virmach.com/ also appears down, but I'm not sure if it's on the same host.
@JBB - You're not alone as it seems that all of Chicago DC #1 (CHIZ001, CHIZ002, CHIZ003, CHIZ004) has dropped. I'm on CHIZ003 and am getting the same "Operation Timed Out ..." error as you are. I expect it may be a DC network issue, or maybe someone unplugged the cable to VirMach's rack because you know VirMach does not have the best luck when it comes to DC hands.
LES • About • Donate • Rules • Support
Longing for a Green Tag
Yo, join our premium masochist club
Thanks, I suspected as much.
It's been a reliable location for me so far, but I have noticed that under interactive use sometimes it feels like the network is a bit intermittent. I'll type commands in my term and nothing for a second or two, then it catches back up. I've never been able to measure it as packet loss, it's both more subtle and more annoying than that.
I guess this means the VPS of mine that's vanished is in Chicago, then
(doesn't seem to be a way to see what node you're on once it's dead... /sigh)
Yes, offline at 12:00 noon Shanghai time
Yes , Chiz is offline.