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
Or just add the flag --no-check-certificate to wget.
@0xdragon that wouldn't be safe. Plus there were a few others with various dependency issues
Not saying it's the right way to do it, just the easiest. I completely agree with you that the installer should do more to check for and install required packages.
Im not sure if someone already asked or i could not find it but i would like to know if there are any plans to make some servers public so everyone with the link can see your servers and the data...
Another question: Have you already a price level in mind?
its already asked for many times. just wait and see..
@Colm
Nice find.
Look at line 274 https://github.com/nodequery/nq-agent/blob/master/nq-agent.sh
thanks i will try it.
i link,used
That may be true for sensitive information but I actually don't consider any of the transferred metrics sensitive right now.
The agent script uses '--no-check-certificate' which really is the only way to use SSL without complications. As for the dependency check, apt is not required for the script and wget is used to download the script so it should be installed anyway. I will look into additional error reporting for the installer but every other tools should be installed on a basic linux system.
Thank you all again for your feedback. We hope to release another update to the agent soon.
From what i remember it depends on how good your cert is. Cheap certs are cheap. They work in most cases but some edge cases they wont work you may need to pony up for a cert that everyone likes.
I expected that from our Comodo certificate (that's why I added the parameter) but definitely not from a company like Github. I do however have to say that none of our test environments gave a warning regarding their certificates.
I dont think that you will see it with most of user base. For example maybe windows server may not like the cert. Maybe redhat may not like the cert or a BSD but they are currently out of scope. One thing you can do is use cloudflare to proxy your stuff and they use certs that should not have any problems.
@Joe_NQ awesome work, finding it quite useful.
I've noticed on a couple of boxes that cron emails are being generated due to stderr writes in the script.
Not a big deal, but managed to chew through a lot of sendgrid credits pretty quickly. Easy to fix by creating the /etc/os-release file manually on affected boxes.
Also is there any way to increase the server limit from 10? I've got 18 nodes to monitor
What kind of pricing are you looking at?
Thank you, that could be an option and I'll look into it.
It is already on our list. We will validate if the release file is available and redirect any cron output to a log file so you won't receive local emails anymore.
We don't know any details regarding the pricing yet. We hope to offer 10 servers for around $5 to $10. I've raised your limit to 20 for now
That pricing sounds pretty good for me :-).
This looks excellent. Great website as well. I'd be interested to know how you're storing the data from a speed point of view, as I'm guessing your database will get very large very quickly.
wich pricing? i didnt see it.
edit this one.:
@Joe_NQ: ...hope to offer 10 servers for around $5 to $10.
Fix it. You shouldn't be not checking the certificate
That's fantastic, cheers.
Great service, added 4 servers, had some trouble with one of them not accepting the github ssl but changed a line in the install script and it worked.
Excellent look and feel -- I'll be trying this out as well.
Hello @Joe_NQ i have some questions:
Yesterday i receive an email one of my servers where not responding, good, but didnt receive an email server is back, should be one isnt?
and the other, in the Manage Servers page
--You have have 9 of 10 servers in your account--
is there a way to arrange servers by name, or ip, or just move them to fit any of the places. like the stats box and graphic boxes inside the info of the server.
thanks.
Until now we are fine with regular caching. Because you won't have the full metric resolution for every interval I don't think size would be a problem anytime soon. However, we would close the registration as soon as things slow down.
As you might have missed this is even happening with Github. On some servers it is simply not working out of the box. Because the metrics are not considered highly sensitive, it is not a priority right now but definitely on the list for future updates.
Thank you, we will consider a notification for servers that come back online. Manual sorting for the servers could be an option too, I will look into it. Until then you could always add numbers in front of their names to sort them.
Thanks to everyone for your feedback. Highly appreciated.
Also @Joe_NQ
-the statistics wold be nice to see it weekly, or not many users ask for them?
-some of my servers shows the latency always is 0.00 can be the firewall? i have some servers around the globe, if you want to test.
-maybe add monitor for common ports? 21, 22, 53, 80, XX, XX, green or red light.
Interesting, seems like the server to test Europe latency is hosted in Italy, as my prometeus VPS has super low latency on the graphs:
A public reports page would be nice by the way!
My guess is that your in the same datacenter. I would not be shocked if your on the same rack or server.
@Joe_NQ This is great and a service I would paid for. Thank you!
Weekly metrics are included in the monthly view, in my opinion it is sufficient for now. Regarding the network latency, maybe you are in the same datacenter as our ping nodes? Or is it showing 0.00 for all three locations? Port monitoring is currently not on the list as we would like to implement process checks first.
Indeed, we are currently hosted with iwStack @prometeus
Thank you, much appreciated.
Second downtime alert that came as an error, it isn't that accurate.