Java Garbage Collection Tuning - -XX:GCTimeLimit and -XX:GCHeapFreeLimit not working? -


in our gc logs, seeing sort of pattern extended periods of time (20+ minutes). here small snapshot of full gcs running during time:

2017-08-17t17:40:55.958-0500: 2263.468: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698843k->698835k(698880k)] 899035k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3358980 secs] [times: user=12.79 sys=0.03, real=0.33 secs]  2017-08-17t17:40:56.294-0500: 2263.805: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698835k->698835k(698880k)] 899027k->899027k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.4146470 secs] [times: user=14.99 sys=0.04, real=0.42 secs]  2017-08-17t17:40:56.710-0500: 2264.220: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698835k->698835k(698880k)] 899027k->899027k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3549330 secs] [times: user=14.16 sys=0.03, real=0.35 secs]  2017-08-17t17:40:57.065-0500: 2264.576: [full gc [psyounggen: 200191k->200191k(272896k)] [paroldgen: 698835k->698835k(698880k)] 899027k->899027k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3333880 secs] [times: user=12.86 sys=0.04, real=0.34 secs]  2017-08-17t17:40:57.399-0500: 2264.910: [full gc [psyounggen: 200191k->200191k(272896k)] [paroldgen: 698843k->698835k(698880k)] 899035k->899027k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3361300 secs] [times: user=13.10 sys=0.04, real=0.33 secs]  2017-08-17t17:40:57.736-0500: 2265.246: [full gc [psyounggen: 200191k->200191k(272896k)] [paroldgen: 698845k->698835k(698880k)] 899037k->899027k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3165070 secs] [times: user=11.66 sys=0.03, real=0.32 secs]  2017-08-17t17:40:58.053-0500: 2265.564: [full gc [psyounggen: 200191k->200191k(272896k)] [paroldgen: 698835k->698834k(698880k)] 899027k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3860970 secs] [times: user=15.29 sys=0.09, real=0.39 secs]  2017-08-17t17:40:58.440-0500: 2265.951: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698834k->698834k(698880k)] 899026k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3679940 secs] [times: user=14.06 sys=0.04, real=0.37 secs]  2017-08-17t17:40:58.809-0500: 2266.319: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698834k->698834k(698880k)] 899026k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3336190 secs] [times: user=12.92 sys=0.03, real=0.33 secs]  2017-08-17t17:40:59.143-0500: 2266.654: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698834k->698834k(698880k)] 899026k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3498820 secs] [times: user=13.31 sys=0.04, real=0.35 secs]  2017-08-17t17:40:59.494-0500: 2267.004: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698834k->698834k(698880k)] 899026k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3055050 secs] [times: user=11.00 sys=0.04, real=0.31 secs]  2017-08-17t17:40:59.800-0500: 2267.310: [full gc [psyounggen: 200192k->200191k(272896k)] [paroldgen: 698834k->698834k(698880k)] 899026k->899026k(971776k) [pspermgen: 26732k->26732k(26880k)], 0.3235970 secs] [times: user=12.22 sys=0.05, real=0.32 secs]

so have full gc running multiple times per second, leaving little time jvm real work. tried setting following arguments force "gc overhead" oom when occurs:

-xx:gctimelimit=5     -xx:gcheapfreelimit=50

here oracle documentation on these reference: https://docs.oracle.com/javase/8/docs/technotes/guides/vm/gc-ergonomics.html

i think these settings in place (>5% of time spent in gc, <50% heap free), triggering outofmemoryerror: gc overhead limit exceeded. thoughts on why not triggering expect?

edit - little more info, here top of gc log, showing version info, arguments, etc:

java hotspot(tm) server vm (24.101-b14) linux-x86 jre (1.7.0_101-b14), built on mar 31 2016 05:26:20 "java_re" gcc 4.3.0 20080428 (red hat 4.3.0-8) memory: 4k page, physical 131954348k(1343412k free), swap 125812704k(125512524k free) commandline flags: -xx:gcheapfreelimit=50 -xx:gctimelimit=5 -xx:+heapdumponoutofmemoryerror -xx:heapdumppath=/cerner/core_files -xx:initialheapsize=446693376 -xx:maxheapsize=1073741824 -xx:onoutofmemoryerror=/usr/local/cwx/ejs_oom.ksh -xx:+printgc -xx:+printgcapplicationstoppedtime -xx:+printgcdatestamps -xx:+printgcdetails -xx:+printgctimestamps -xx:+printheapatgc -xx:+useparallelgc

one possibility have disabled -xx:-usegcoverheadlimit

edit - little more info, here top of gc log, showing version info, arguments, etc: -xx:gctimelimit=0

that's not option have set.


Comments

Post a Comment

Popular posts from this blog

Is there a better way to structure post methods in Class Based Views -

i'm using class based views in django , never specify url in ajax calls , omit action parameter in form because know these requests go through post method of class based view. notice view - profile page make many requests depending on user updating on page - post method become convoluted , if/else mess example: def post(self, request, *args, **kwargs): # handle user changing available date if request.post.get('availabledate') != none: self.updateavailabledate(request) return jsonresponse({'result':'success'}) if request.post.get('newprojectname') != none: creative_user = creativeuserprofile.objects.get(id = request.user.id) project = project.create_project(creative_user, request.post['newprojectname']) project.save() return jsonresponse({'projectid': project.id }) if request.post.get('projectimage') != none , request.files['file']: proj
Read more

performance - Why is XCHG reg, reg a 3 micro-op instruction on modern Intel architectures? -

i'm doing micro-optimization on performance critical part of code , came across sequence of instructions (in at&t syntax): add %rax, %rbx mov %rdx, %rax mov %rbx, %rdx i thought had use case xchg allow me shave instruction , write: add %rbx, %rax xchg %rax, %rdx however, dimay found agner fog's instruction tables , xchg 3 micro-op instruction 2 cycle latency on sandy bridge, ivy bridge, broadwell, haswell , skylake. 3 whole micro-ops , 2 cycles of latency! 3 micro-ops throws off 4-1-1-1 cadence , 2 cycle latency makes worse original in best case since last 2 instructions in original might execute in parallel. now... cpu might breaking instruction micro-ops equivalent to: mov %rax, %tmp mov %rdx, %rax mov %tmp, %rdx where tmp anonymous internal register , suppose last 2 micro-ops run in parallel latency 2 cycles. given register renaming occurs on these micro-architectures, though, doesn't make sense me done way. why wouldn't register renamer
Read more

jquery - Responsive Navbar with Sub Navbar -

good day, forgive me nav vocabulary not best. if question confusing apologize. i'm working on navigation bar needs centered on page @ 768px , above. items in nav open (or perhaps toggle) subnav lives directly below. these nav should appear centered on page. i have been able work through responsive portion of nav bar, having main items align vertically , sub nav menu of each item roll out below, , gave them indent separation. working well. however, issue not responsive side, behavior @ width greater 768px. happens toggling menu's sub navigation pushes other main items down page below sub nav. here's code: $(document).ready(function () { //stick in fixed 100% height behind navbar don't wrap $('#slide-nav.navbar-inverse').after($('<div class="inverse" id="navbar-height-col"></div>')); $('#slide-nav.navbar-default').after($('<div id="navbar-height-col"></d
Read more