nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

All the time GC Allocation Failures

Tomasz Dubiel_pc5973

Hello. This is the usage of RAM by Java on my Jira server. Plenty of RAM is available. 8 GB of RAM, free RAM 6 GB, RAM used - 2 GB. However, I see all the time allocation failures in Jira logs.

2021-09-01T08:38:03.148+0200: 51009.205: [GC (Allocation Failure) [PSYoungGen: 2071466K->24389K(2066432K)] 3301838K->1255115K(8357888K), 0.0756700 secs] [Times: user=0.14 sys=0.00, real=0.08 secs]
2021-09-01T08:38:27.204+0200: 51033.239: [GC (Allocation Failure) [PSYoungGen: 2065221K->21547K(2068992K)] 3295947K->1252671K(8360448K), 0.0756685 secs] [Times: user=0.15 sys=0.00, real=0.07 secs]
2021-09-01T08:38:40.282+0200: 51046.308: [GC (Allocation Failure) [PSYoungGen: 2062379K->22544K(2068992K)] 3293503K->1254392K(8360448K), 0.0699382 secs] [Times: user=0.15 sys=0.00, real=0.07 secs]
2021-09-01T08:38:58.290+0200: 51064.300: [GC (Allocation Failure) [PSYoungGen: 2063376K->21217K(2068992K)] 3295224K->1254177K(8360448K), 0.0699216 secs] [Times: user=0.14 sys=0.04, real=0.06 secs]
2021-09-01T08:39:21.889+0200: 51087.884: [GC (Allocation Failure) [PSYoungGen: 2062049K->22551K(2069504K)] 3295009K->1256084K(8360960K), 0.0839347 secs] [Times: user=0.17 sys=0.02, real=0.08 secs]
2021-09-01T08:39:33.192+0200: 51099.175: [GC (Allocation Failure) [PSYoungGen: 2064407K->21237K(2069504K)] 3297940K->1256218K(8360960K), 0.0781307 secs] [Times: user=0.15 sys=0.00, real=0.08 secs]
2021-09-01T08:39:44.029+0200: 51110.003: [GC (Allocation Failure) [PSYoungGen: 2063093K->23253K(2070016K)] 3298074K->1259246K(8361472K), 0.0635178 secs] [Times: user=0.16 sys=0.00, real=0.06 secs]
2021-09-01T08:40:04.280+0200: 51130.237: [GC (Allocation Failure) [PSYoungGen: 2066133K->19490K(2070016K)] 3302126K->1256151K(8361472K), 0.0752106 secs] [Times: user=0.16 sys=0.01, real=0.08 secs]
2021-09-01T08:40:17.420+0200: 51143.368: [GC (Allocation Failure) [PSYoungGen: 2062370K->20774K(2070528K)] 3299031K->1257788K(8361984K), 0.0600858 secs] [Times: user=0.14 sys=0.00, real=0.06 secs]
2021-09-01T08:40:29.256+0200: 51155.194: [GC (Allocation Failure) [PSYoungGen: 2064166K->22946K(2070016K)] 3301180K->1262055K(8361472K), 0.0653567 secs] [Times: user=0.17 sys=0.00, real=0.07 secs]
2021-09-01T08:40:38.598+0200: 51164.528: [GC (Allocation Failure) [PSYoungGen: 2066338K->16663K(2069504K)] 3305447K->1256298K(8360960K), 0.0960412 secs] [Times: user=0.19 sys=0.04, real=0.10 secs]
2021-09-01T08:40:53.697+0200: 51179.616: [GC (Allocation Failure) [PSYoungGen: 2059543K->23105K(2070016K)] 3299178K->1262836K(8361472K), 0.0859570 secs] [Times: user=0.14 sys=0.02, real=0.09 secs]
2021-09-01T08:41:05.113+0200: 51191.023: [GC (Allocation Failure) [PSYoungGen: 2065985K->18449K(2070016K)] 3305716K->1258365K(8361472K), 0.0762517 secs] [Times: user=0.18 sys=0.01, real=0.08 secs]
2021-09-01T08:41:13.169+0200: 51199.071: [GC (Allocation Failure) [PSYoungGen: 2061329K->23151K(2070016K)] 3301245K->1263696K(8361472K), 0.0815816 secs] [Times: user=0.15 sys=0.00, real=0.08 secs]
2021-09-01T08:41:22.953+0200: 51208.847: [GC (GCLocker Initiated GC) [PSYoungGen: 2066031K->20116K(2070528K)] 3306579K->1264313K(8361984K), 0.0631541 secs] [Times: user=0.13 sys=0.02, real=0.07 secs]
2021-09-01T08:41:34.448+0200: 51220.335: [GC (Allocation Failure) [PSYoungGen: 2063508K->15170K(2070016K)] 3307706K->1259550K(8361472K), 0.0779939 secs] [Times: user=0.15 sys=0.00, real=0.08 secs]

What is the reason of these errors? We use JIRA Service Desk 3.16.11, Jira Software 7.13.11 and Jira Core 7.13.11.

Accepted answers

All comments

Mirek

Hi @Tomasz Dubiel ,

Allocation errors might be caused by few things but it requires a little bit more information what is exactly going on at that specific moment, which threads are working and how the server memory is used (not only Jira memory). Thanks to that you can tweak your memory settings to be sure that everything is working fine.

I remember @Andy Heinzer investigated and explained pretty well on a similar topic

https://community.atlassian.com/t5/Jira-Software-questions/Garbage-Collector-Alllocation-Failure/qaq-p/1322269

I recommend to read everything and check whatever Andy is suggesting.

You might be also hitting some of those problems

https://confluence.atlassian.com/jirakb/jira-running-out-of-memory-due-to-gc-allocation-race-condition-957122851.html

https://confluence.atlassian.com/jirakb/jira-becomes-unresponsive-occasionally-and-logs-several-full-gc-events-1014679965.html

However your pause time does not seem to be long, so I guess application if fully responsive on those moments.

Tomasz Dubiel_pc5973

Yeah, application runs smoothly, we don't experience any problems, but these errors are just spamming in log.

Should we try to give Jira minimum and maximum RAM usage at 4GB and increase RAM on server to 16 GB? The total RAM on server is now 11,5 GB. I've read in topic
"We always expect the system to have about 4x the system memory as the heap Jira will use"

Tomasz Dubiel_pc5973

Although, indeed, for some users Jira becomes unresponsive.

Mirek

Yes system should be also having enough memory to run smoothly. Check also your Xmx and Xms parameters. It is nice to have them on the same value.

Daniel Ebers

Hi @Tomasz Dubiel

first of all the sad news that debugging something like you are seeing is pretty hard from the outside - without knowing the system in-depth and without having access to it's console.

That being said and adding to the great information you have already received two things come to mind:

1.) The version of Jira (7.13) is rather old and meanwhile unsupported. Upgrading, if possible, to a more recent version could lead to improved performance and a better experience (for users but also for admins)

2.) I seem to recognize that you are not using the g1gc garbage collector. You could double check in your configuration file "setenv.sh":

While I do not want to say this is the golden key for whatever - we saw in the past a better management of memory when switching to that garbage collector.

You can read more about it here:

Garbage Collection (GC) Tuning Guide
https://confluence.atlassian.com/enterprise/garbage-collection-gc-tuning-guide-461504616.html

Please note the prepending sentence which might lead into a solution (but that is no guarantee!):

In many circumstances, we now recommend the Garbage First Garbage Collector (G1GC).

Please test it, if possible, on a test environment first. Currently I do not recall if this recommendation was all valid for 7.13 versions already - for a 8.x version it is confirmedly a good advise.

Regards,
Daniel

Welcome!

It looks like you're new here. Sign in or register to get started.

Quick Links

Find more posts tagged with

server

jira

jira-server

garbage-collection