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
Agree, I am with them for 3 years,
https://docs.google.com/spreadsheets/d/1S9sUGyNEOnrMYsmmkxb8NkVe-0MBICqAoyPyrwft0UI/edit#gid=0
NOOOOOOOOOO DONT SHOW THAT
PTSD flashbacks
a) so much missed
b) so much money wasted
c) so much money could be saved
d) VirMach active on LET :<
Wow, speechless.
Not my credits, someone made it.
in memoriam Sir
The stage is set:
I just want the server's IPMI information "E3-HD New York"!!!
For GOD Sake!!!!
We have hundreds of people contacting us about the web VNC not working, after we put up a pop-up modal and help article regarding the issue. They contact us with different titles and describe the issue in different vague ways, making it difficult to go through all them and slowing down support in all other cases.
You need machine learning to automatically identify and answer the tickets, instead of blaming customers for creating too many different tickets in vague language.
You also need to pay for enterprise level on Cloudflare so that WebSockets will work, or move WebSockets to a separate subdomain not using Cloudflare, or reprogram noVNC using the new QuicTransport that might perform better.
Yeah, with their margins, I'm sure a major investment in AI to help idiots is warranted.
Now, now ... maybe he meant for the response to be, "Learn to read. Ticket resolved."
It's basically a keyword match to identify similar tickets. IBM Bluemix can do that easily. Paying $1000 monthly and hiring one fewer personnel that would cost $4500.
Customers have to go through a modal that pops up on the page where they can launch the web VNC. This would mean they would have to actively ignore this message. In addition, when they create a ticket, in most cases they would be presented with a knowledgebase article before they create the ticket as well. Machine learning would not do a terrific job and even if it's sorted perfectly, we would still have too many tickets as we didn't expect this quantity of tickets after explaining the situation in multiple places.
Someone mentioned their ticket taking a long time to receive a resolution. I was simply explaining why, but if you see this as us blaming customers for creating these tickets (at least those with limited support packages) then that's fine as well.
We already have the correct plan.
We're already swamped with projects and an influx of tickets due to Black Friday renewals. Unfortunately, that doesn't leave any time for that.
I'll explain to you why this would not work.
As I mentioned, we have people creating these in such ways that it is vague -- this means that it could be other potential issues. AI would not know whether it's that other issue or this one when it's described to us completely wrong. The only reason we're able to answer the majority of them is that we can make educated guesses based on the exact situation, and factoring in a lot more than just keywords. We look at signup dates, how many services they have with us, how long they've been a customer, whether they provided VNC, SSH, or RDP details, and so on.
Then, the reason this is being highly problematic would be that the type of person that ignores a notice regarding this temporary issue also ignores our responses. They continuously make multiple tickets and do not provide the details required to assist them. They do not correctly read our responses. They confuse basic terms such as VNC and VPS, and try entering VNC details into RDP and so on. AI would definitely not be able to handle this well and once it is sorted, we would still have to go back and forth many times to explain the same thing. If we cannot resolve these quickly and many tickets exist, there's not much we can do. On top of this, there are customers with mindsets such as yourself which use this as an opportunity to argue with us or ask us to things differently. Then, there's those who complain and act as if their service is completely inaccessible because they temporarily have some convenience removed and instead of one click, it is one click, three copy/pastes, and a download. I understand in some cases people do require the web VNC and we're trying to work with people to provide a temporary solution.
I just tried with my account.
I can see the modal in WHMCS "active product" section and a notice on the "submit ticket" page.
If I login directly to SolusVM, there's no notice about web VNC problem.
The "address" and "port" given are correct, but it would be nice if you add "display" number (
display = port - 5900) that is required bygvncviewer.Charge $3 if someone creates a second ticket within 24 hours.
If the second ticket is genuinely a different problem, refund the $3 to account credit.
Take your time, this is the fourth day!!!!
If the second ticket is genuinely a different problem, refund the $3 to account credit.
We've had this idea in the past, it wouldn't work. We'd get even more tickets/upset people if we charge deposits before doing work. This is something we generally do with unmanaged dedicated servers, where we charge $15 to troubleshoot issues and refund it if it's an actual problem with hardware or networking because we get a lot of people that change their networking configuration/break it, and expect free assistance.
We've switched over to the majority of users using the control panel button to sign in to SolusVM, but you are correct if they exclusively use SolusVM. The majority of these requests are not related, they are newer orders where the customer is utilizing the VNC button from WHMCS.
This is a temporary measure that works with the majority of popular VNC applications. I'm not sure if SolusVM outputs that on their API, but I'll have a look as soon as we get these ticket quantities in control.
You should most likely receive a reply today. I've been merging duplicate tickets regarding the same issue so it may have been merged, causing a delay but I'll try to remember your name and get to this sooner.
I noticed that this was on a new order so we'll most likely adjust your due date. Alternatively, if you would like in your case we can offer a refund/cancellation as another choice should you decide to go elsewhere. Let me know if you prefer that before we get to your ticket and we can process it that way instead.
No I don’t want to cancel it, please just reply to the ticket with needed info.
And thank you for adjusting due date..
While they do work, the big issue with 3rd party VNC viewers is the lack of an encrypted connection.
If you can get the port number from SolusVM API, the display number is simply
port - 5900.VNC fundamentally isn't encrypted. The WebSocket VNC app encrypts between your browser and the web server, but traffic between the web server and the VNC server is still in the clear.
When I setup a box via VNC, I enter a temporary password, and then change the password over SSH.
Good security practice from the guy who keeps his laptop password written on a sticker attached to the bottom
Why are you laughing? That's a great idea! I tried that, but kept forgetting where I put it. So, now I use my desktop wallpaper to save all of my passwords. The laptop password is engraved onto the screen bezel, of course. I would hate to lose it.
My password is posted on this forum. Don't transfer my money. Honor system. Thanks.
https://www.lowendtalk.com/discussion/comment/3134457/#Comment_3134457
@virmach but for the fact that your firm and/or your vendors lack the technical skills/time/inclination to fix what users perceive as a broken system (because it worked a long time using a certain approach), you wouldnt be inundated with tickets. But for the fact that your run a staff that apparently is too lean, you wouldnt have a huge backlog of tickets. I know I know, those pesky customers never phrase things exactly the way your tier 1 needs to comprehend things... those pesky customers, blah blah blah.
In this case, but for the failed system (meaning it worked before and is broke) and a failure in communication - the ticket typhoon is entirely of your creation. Then you try to sell new stuff and can't give a client credentials in 4 days but would take time to refund said client? We both know what that means- your vendor is backlogged and hasn't provisioned your machine so you have no credentials to give the client, for if you did- you'd send credentials and not offer refund.
Hence all your rambling about customers is best kept to yourself- you created the typhoon.
I imagine the hardware has to be available for it to be in stock. You're making a pretty big assumption that the vendor is the bottleneck when it's clear they have a bottleneck with human and manual actions. Very few providers are fully automated on dedicated servers. And it's not like there's a global pandemic going on, affecting staffing levels or anything.
They have a good chunk of non-English speaking clients. We've seen lots of examples of the kind of shitty, intelligible tickets Virmach is talking about.
You're coming off as irrational and butthurt. If you did one day of tech support in your life, you wouldn't be talking this way.
You can totally be pissed for having slow support and taking on more business then they can handle, but that's almost all businesses that are trying to grow.
It sounds like Terms of Service agreements should be replaced with simple technical skill testing questions so the completely helpless support drainers are turned away.
There's a difference between not describing something well and ignoring notices that are placed in a position where nearly everyone should see them. This includes network outages, if we post it on the outage page and update it there then customers should know to check this first before opening a ticket when they are specifically instructed to take a look there first if their service is down.
We can spend more time on the issue, which leaves less time for the tickets. It's a balancing act and no matter what, some people will be displeased.
There's also a difference between not having the technical skills to resolve an issue and not being the ones who developed the system. Nothing we did caused this to break, or should have caused it to break. It was not included in the changelog for the software, nor were any changes described/announced by Cloudflare. This could be compared to the time Cloudflare broke our captcha pages by not announcing anything at all, not communicating the change to their staff, and not coding their change properly (something we do not have full access to change.)
Our "lean" team works out as long as the workload does not increase 5-10x. In this case, it's caused temporary delays in responses. We cannot keep 5x the staff members just in case one time in multiple years this occurs.
I'm interested to know how you believe this could be communicated better so we can implement the change. We could place the notice on the VNC page itself as well, but if they skipped reading it before they clicked the VNC button, I'd assume it is skipped on the VNC page as well.
The credentials are provided. We quickly added this in. Should they not be provided directly on the billing panel, it would still have also been available on the full control panel page, as are many other features not prominent on the quick access version. This has been the case for years. You can go to the control panel, click on VNC, and see the VNC IP address, port, and password.
This is completely unrelated to the issue.
And once again, I believe it's important to state again: we were only replying to someone on why there is a delay in their ticket being resolved. It is an explanation. You do not have to agree with it, but it is what caused the delay.
Since this seems to be going in the direction of dedicated servers, I do want to just state that some other issue is causing the backlog and all tickets are being handled more slowly than optimal.
The dedicated servers are mostly deployed and we've deployed them mostly within the normal timeframe, it's just that this specific customer requested IPMI information which is not delivered by default, and it's taking way longer than usual because the ticket is stuck somewhere between a lot of other tickets. That's all.
will there be rotating offers similar to that of last year?
Oh no, 384 MB RAM, that reminds me the $1 deal and its 100MB RAM. It can barely install anything.
Just waiting for flash sales.
I can setup nginx with php-fpm7.4 and mysql 8 on such a box.