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.
Understanding Intermittent Outages of vps
I have a website running on vps from greencloudvps and I am working to monitor it via HetrixTools.
In recent days I have started to receive notifications via Telegram that the website has stopped. I am trying to open ssh but I cannot connect. I am trying to browse the website but I cannot also. After a few minutes the website comes back online. I open ssh and send a "htop" I see that the load on the server is very low and that the uptime is 30 days
Also, the site works via cloudflare and I do not see any DOS attacks
What is the possible cause ?
Comments
Maybe the provider is having network problems?
@aaxaa out of curiosity: what location ?
San Jose, CA Location
Newtork
Possible cause could be network issues or server configuration problems leading to intermittent downtime. Check server logs for more insights.
It may due to network problems of the node (perphase DDoS attack). From my experience, if any of the VPSes in the same node get DDoSed, the rest of the VPSes will be impacted as well, which is true to most providers except big players like AWS, Digitalocean. You may want to monitor some other IPs within the same /24 subnet of your VPS using Hetrixtools.
As a provider I can say it could be a network problem at the router. I have seen a router have too many connections. Also the host machine could be overloaded and cause a vps to stop responding for a few minutes.
Probably you will not be able to determine anything from within your vps if the cause is not within your vps. Ask the provider and hopefully they will be honest about what is wrong.
Ive been having the same issue with the San Jose location. Everytime i open a ticket with GC it comes back online but they say they dont find anything wrong
This is how frequently i get those offline @NDTN Hope you can prompt your team to actually investigate this.
@aaxaa I was able to locate your ticket with us via the subject. As our staff replied, we have not seen any issues with the network in SJC or the node issue. Uptime is solid there, we have our own monitoring via Uptimerobot with 1 minute interval - no incidents reported, load of the node is normal (~20), plenty of RAM left (~200GB). More information needed if you need us to investigate further, like tracert/MTR to the VPS when it's down, ping from other sources like our site ping.pw or ping.pe, check from the VNC Console from our Client Area/SolusVM Control Panel.
@NDTN
Thank you for your excellent servers. I have many of your servers and they are all more than good. I have been a "greencloudvps customer" since 2022 and I have never suffered from a problem except for the only problem above, and because it is somewhat strange, I wrote this topic here to see opinions and previous experiences about it.
Thank you also for following this topic and commenting here
Well, their status seems a bit misleading. At least one of their node in Singapore DC1 went down for about 15 minutes almost 12 hours ago, yet their status page displayed that everything was functioning normally. Additionally, their New York node(s) have some problems several hours ago, but the status page indicated that all systems were operating smoothly.
I very often have to investigate weirdest things where logs don't really help. That kind or remark always reminds me of that cartoon.
This happens often for us too. One customer reports small outages during a day or more.
In most cases it is a routing issue that resolves itself, a provider or another is having some route failing, changing to a back-up, then back on. If the connection is flapping there, it can happen more than once (2 outages). If the server is accessible form another location (which is often the case), then that is most likely the issue.
There is no misleading here. Our status page is for monitoring the whole network of the DCs. If one node goes down, it does not mean that there is a network issue in that location. There was a maintenance in Singapore DC1 on the mentioned node yesterday, to replace a bad RAM stick if I'm not wrong, you should have received a maintenance email already. For the NY node, we are investigating a possible issue with the kernel.