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.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
@rchurch genuine question: to BOOT from an ISO why use the REINSTALL function instead of selecting it in the DISK EDIT section just shown? (obviously choose it as 'boot device 1' as well)
sorry, I am not a customer, so only can judge by the screenshots and not try myself ;-)
I’m aware what you mentioned but this different story.
@key900 This is what you issue is. You are depending on some of your own customers who think I am picking on you without justification to prove that your system has bugs and they have proved it, and so have I with the last post with showing that
parted -l
cannot detect the partition information. This happened on two different hosts. cl10 and cl14.Now you have gone and terminated the monthly nodes before they are due, because they are still proving that the problem exists. I am certain that a workaround can be found for the problem until the underlying problem is fixed, but you won't play ball and you prefer to sweep the problem under the rug.
If that is the case, then you'd better refund them.
Instead of admitting that there may be an issue you and your incompetent tech staff blame the customer. I'm sure most of the people in this part of the thread are 10 times more knowledgeable than your own tech staff, one of whom appears to be a bot.
So @rchurch is trying to boot from the block device?! (So this is what all the fuss is about?)
No it is a simple matter. Partition info on the NVMe should not be lost when you are reinstalling from an ISO unless you expressly overwrite the partition. You can boot from the block storage if grub is properly configured but you shouldn't lose the NVMe information no matter what.
Failure to meet a reasonable expectation is a bug
Inadequate documentation is a bug
Faulty documentation is a bug
Problems which happen as a consequence of not following established convention is a bug.
Blaming the customer without justification is a bug
If booting from an ISO requires editing the disk boot order rather than reinstalling an ISO that is fine, but where is that clearly documented?
Why should a customer think that a disk with "rescue" in the name or a "rescue mode" in an installation ISO will destroy the NVMe partition info, when that is the job of a rescue disk or a rescue mode to fix its problems?
Since you have deleted my access to the system, what matters now is if someone else with access to the system can verify whether booting an ISO by editing the disk order, rather than Reinstalling with an ISO will leave the partition information of the NVMe intact, and test the process across all hosts.
Then you can improve your documentation and decide whether the Reinstalling with an ISO option should be disabled or not.
Providing a service is a process, not just a technical matter. Bugs are not just technical issues.
Do you remember the Samsung Galaxy Fold debacle when Marques Brownlee and some other reviewers peeled off what they thought was a screen protector not knowing that it was part of the screen itself?
That was a bug on Samsung's part. The customers where acting based on prior experience and expectation, regardless of what the documentation or whatever said.
The ball is now in your other customers hands.
This a part what he doing. Last time i saw it he installed grub and /boot / swap at Block storage then tried to boot from Main disk his system. Then last swapped and tried to boot from Block Storage.
First claims that he formatted the main disk and when he rebooted the formatting is gone! Did you ever saw that anywhere? It quite seems he don’t even know how to formatting his disk.
LOL!
is that looks like someone know how to dealing with formatting?
@key900 and @dahartigan
At least experienced technical people can see the kind of clowns who run Letbox and the idiots who come to defend (applaud) their incompetence.
Do you recognize the implications of this screen?
This screen should tell you that Proxmox sees all the disks as block devices and expects to be able to boot them if they are properly configured. This Proxmox/Seabios is the system you are using in case you need reminding.
It is a terrible thing for your own mouth to be your own footgun. Please stop this, the anguish I'm feeling for you is unbearable.
I will tell you this story about a dude who was on trial for murdering and quaffing a vulture. (some people do eat vultures for real). When the accusations were laid at him, he addressed the court. "Okay, so you guys say I murdered and wolfed down a whole vulture. What evidence have you got? If you guys say you saw me, tell me, which hand did I use to kill the vulture, my left hand or my right hand", gesticulating emphatically with his arms, one after the other. So the audience yelled "you used your left hand!!"
Then he responded "Aha, you are all wrong. It is my right hand I used!!".
There in lies the moral of the story. Enough said. Just fix your stuff and stay out of LET comment threads.
TLDR anyone?
@rchurch claims to have found a bug, but so far it's a bug that only he has experienced and is able to reproduce.
Although this isn't a totally improbable scenario, it's odd that others haven't (yet?) experienced this bug.
Others may have found the bug, but Letbox tech support probably told them it was their fault and they believed it, or it probably just wasn't worth their while to rant on LET about it.
You yourself mentioned that your block storage had vanished when you booted it after a long period of idleness. Don't you think you would have lost data if you had actually stored data on it?
My main issue is that their tech support are clueless as @key900 himself has repeatedly proven on this thread, and you'd better be wary if you intend to store your data with them.
I honestly don't know. (In my mind, I took it to be more a question of whether the disk appeared in the control panel or not, but I can't say for sure. In any case, as I said above, it seems that no one has yet reported lost data.)
As for your issue: maybe. I'm trying to remain neutral -- my main objection earlier was to your cursing and shouting.
I was getting annoyed with @dahartigan insulting from the sidelines without knowing the facts.
The issue is a simple one. If a person installs with a template first, then boots via Reinstallation with an ISO, will the NVMe partition have been erased even before they start making changes or not?
I have confirmed it more than once on different VMs and posted the info to show it.
It is a simple matter and neither @key900 or his self-appointed sidekick @dahartigan have refuted it after all this back and forths. They have simply been beating about the bush trying to distract from this fact.
Reinstallation with ISO
If what this screen says is true then it is a bug, because it is the first place people will go if they want to boot into a rescue disk.
Cue Marques Brownlee video again
I wasn't aware that option existed. Virtualizor's screens are a mess, if that is what they are using.
This is the main point of my complaint
https://www.lowendtalk.com/discussion/comment/3029525/#Comment_3029525
Okay, my last attempt to try to help: if you simply want to boot an ISO, don't use the "Reinstall" function.
Instead: in your client space, under "Information", find "CD/DVD Disc Image File" and (to the right) choose the ISO image that you want to boot, and then confirm. Then (just below) set the value for "Boot Order", and then confirm. Finally, press the "Reboot" button above under "Actions".
If you do it this way, I can't imagine that the mere (re)booting of an ISO would delete anything.
They are good for specs
They had uptime issues and some tech, and probably will
Don't expect much for the price, do your backups, yes 3Tb+
And this is the point I am trying to make about Letbox and @key900.
Why isn't this method explained by key900 and his tech support staff after so many hours?
Why does he and his tech support guys have to come and learn how to administer his own systems from users on LET?
I entered this thread to warn about their clueless tech support. I was already done with the service when I came here.
Please take this the right way, but: wow, this is a bit anticlimatic as an ending.
Setting aside a potential language barrier, and also setting aside that these are unmanaged services, a few of us were also scratching our heads for a good while today about what you might be doing wrong -- it wasn't transparent.
So I wouldn't be so hard on LetBox/@key900. Not to mention that they have a knowledge base that mentions how to mount an ISO.
Sometimes, especially if a control panel is unfamiliar, one needs to study and experiment with the various options offered. This is to be expected.
The escalation that took place in this thread today really wasn't called for.
Tldr rchurch doesn't understand what hes doing, wiped his data and overreacted to the point where most unmanaged providers should just blacklist his buisness.
Here comes another shill pimping for providers with clueless tech support.
Why don't you check the date I registered on LET before talking shit.
I was on LET before most of these providers even thought of selling hosting services and I was administering VMs when most of the current crop of newbie providers were in primary school.
@rchurch you should be calm down now and let it go. You are being too personal now.
and the network is pretty good in the US
And this is the point I am trying to make about Letbox and @key900.
Why isn't this method explained by key900 and his tech support staff after so many hours?
Why does he and his tech support guys have to come and learn how to administer his own systems from users on LET?
I entered this thread to warn about their clueless tech support. I was already done with the service when I came here.
lmao bro chill
Maximum snowflake level attained.
No, lots of others, such as myself and ehab have had the disk order issue. It's not a boot order issue, but disk order issue.
It was frustrating that it took multiple tickets to explain the issue, but was resolved.
The issue is that the Control Panel allows you to select device boot order, but not disk order. Something makes seabios change disk priority to boot the block storage and not NVMe. If bios was exposed like in ESXi, we could fix ourselves, but since it's not, their Support needs to do it. I asked twice what they did and didn't get answered.
Workaround is using noVNC console to press ESC on boot and manually selecting the virtio disk a. But you have to do that every boot until Support fixes it. Ain't nobody got time for that. If I reboot a server, I expect 30-60 seconds downtime, not a hassle of console access.
So when I saw key900 say this isn't a bug and you fix in boot order screen, I nearly lost my shit because he's one or two days behind in knowing this problem is on his side to fix. In my ticket reply, they said they'd report the bug to developer. FFS.
So the language barriers and support responses make me think this is fine for personal use, but not for critical business use.
Yes they may have language barrier, yes they have some templating issues (ipv6 etc) but for their price, they cpu stronk, network solid now with 10G port and a+b power feed.
Good block storage speeds too.