In my previous posts (e.g. here and here) I showed how to use ps output (e.g. from ExaWatcher) visualization to spot performance problems in Linux. Here I’d like to show that this approach can be taken a little bit further, namely, to find the source of increase in memory usage.

The R code for this is quite straightforward. I also think it shouldn’t be much of a problem to do the same in Python, although I haven’t gotten around to try it myself. In the code below read_ps is the function that reads in a ps output file from ExaWatcher without unzipping it, sum_and_tidy does the aggregation, and visualize_rss does the plotting. There is also an auxillary function keep_top_n which is needed to keep the number of color bands to something reasonable.

Continue reading “Where did my RAM go?” →

In my last post, where I analyzed the problems caused by memory fragmentation on a Linux server, I said very little about memory fragmentation itself. I wanted to tell a story, so I had to dial down the technical stuff. But now that the story is told, I figured I should make a more technical post on the subject. I think it might be useful for many Linux DBAs or SAs, since there is not enough awareness of this problem within the Linux community. So if your system is experiencing cryptic performance and stability issues, keep reading, your problem may well be stemming from memory fragmentation.

Continue reading “Memory fragmentation: the silent performance killer” →

Every upgrade is a bit of a lottery. But for a long running, well established system with a lot of legacy code, it can be more of a Russian roulette. You are unlikely to gain much — your system has been running fine, thanks to various tweaks the apps team applied here and there over the years. All these nice and shiny features that arrive with the upgrade — well, sure, there can be a few that you might like and have big plans for, but most of them are not going to be very useful, at least not out of the box. While there’s a good chance that some of them would backfire in a really, really bad way.

Continue reading “How to hang a server with a single ping, and other fun things we learned in a 18c upgrade” →