I am running an API on a linux server. For the majority of the day the API runs completely fine, however every single day at 1:00 am/pm I receive a large spike of failures that the end after about 5 minutes. Looking into the failures, there is not one consistent pattern in the requests, and most requests still process fine (just a higher proportion fail). It also is not likely traffic related as 1:00am is a very slow traffic window

I am not the only person who is using this server though, so I suspect that someone else is running a process every 12 hours at this time which is eating up a lot of resources.

My question is if there is a bash command which I can run on my server to see if there are any processes that are always being run during this window?

CodePudding user response：

Use top command in batch mode to display system statistics in real time and save the output in a textfile

top -b -n 1 > top.txt

To grab more than one iteration of top command, change -n 1 to your disired number. For example -n 100

top -b -n 100 > top.txt

Put the command in a script and execute the script via crontab or as an atjob at 1:00 am then check your textfile