RE: Solr uses lots of shared memory!
Hello Kevin, Rick, These are interesting points indeed. But this thread is about shared memory, not virtual memory. Any value higher than 0 for MALLOC_ARENA_MAX only reduces virtual memory consumption, from 22 GB to 16 GB. There are no differences in shared memory nor resident memory. Although intrestesting, it unfortunately does not answer the question. To answer Rick's question, the difference with MALLOC_ARENA_MAX=2 is less virtual memory but no changes in queries/second. Thanks, Markus -Original message- > From:Rick Leir > Sent: Sunday 3rd September 2017 17:08 > To: solr-user@lucene.apache.org > Subject: Re: Solr uses lots of shared memory! > > Hi all > Malloc has a lock while it is active in the heap. If there is more than one > thread, and malloc finds the lock in use, then it avoids waiting on the lock > by creating a new 'arena' to hold its heap. My understanding is that a > process with multiple threads which are all active users of malloc will > eventually have an arena per thread. If you limit the number of arenas, you > may suffer delays waiting on locks. > > But this needs performance testing. My experience was with C++, not with a > JVM. I would be interested to know if setting MALLOC_ARENA_MAX=2 makes a > difference to performance. > Cheers -- Rick > > On September 2, 2017 1:15:38 PM EDT, Kevin Risden > wrote: > >I haven't looked at reproducing this locally, but since it seems like > >there haven't been any new ideas decided to share this in case it > >helps: > > > >I noticed in Travis CI [1] they are adding the environment variable > >MALLOC_ARENA_MAX=2 and so I googled what that configuration did. To my > >surprise, I came across a stackoverflow post [2] about how glibc could > >actually be the case and report memory differently. I then found a > >Hadoop issue HADOOP-7154 [3] about setting this as well to reduce > >virtual memory usage. I found some more cases where this has helped as > >well [4], [5], and [6] > > > >[1] > >https://docs.travis-ci.com/user/build-environment-updates/2017-09-06/#Added > >[2] > >https://stackoverflow.com/questions/10575342/what-would-cause-a-java-process-to-greatly-exceed-the-xmx-or-xss-limit > >[3] > >https://issues.apache.org/jira/browse/HADOOP-7154?focusedCommentId=14505792 > >[4] https://github.com/cloudfoundry/java-buildpack/issues/320 > >[5] https://devcenter.heroku.com/articles/tuning-glibc-memory-behavior > >[6] > >https://www.ibm.com/developerworks/community/blogs/kevgrig/entry/linux_glibc_2_10_rhel_6_malloc_may_show_excessive_virtual_memory_usage?lang=en > >Kevin Risden > > > > > >On Thu, Aug 24, 2017 at 10:19 AM, Markus Jelsma > > wrote: > >> Hello Bernd, > >> > >> According to the man page, i should get a list of stuff in shared > >memory if i invoke it with just a PID. Which shows a list of libraries > >that together account for about 25 MB's shared memory usage. Accoring > >to ps and top, the JVM uses 2800 MB shared memory (not virtual), that > >leaves 2775 MB unaccounted for. Any ideas? Anyone else to reproduce it > >on a freshly restarted node? > >> > >> Thanks, > >> Markus > >> > >> > >> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ > >COMMAND > >> 18901 markus 20 0 14,778g 4,965g 2,987g S 891,1 31,7 20:21.63 > >java > >> > >> 0x55b9a17f1000 6K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java > >> 0x7fdf1d314000 182K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libsunec.so > >> 0x7fdf1e548000 38K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libmanagement.so > >> 0x7fdf1e78e000 94K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnet.so > >> 0x7fdf1e9a6000 75K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnio.so > >> 0x7fdf5cd6e000 34K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libzip.so > >> 0x7fdf5cf77000 46K > >/lib/x86_64-linux-gnu/libnss_files-2.24.so > >> 0x7fdf5d189000 46K > >/lib/x86_64-linux-gnu/libnss_nis-2.24.so > >> 0x7fdf5d395000 90K /lib/x86_64-linux-gnu/libnsl-2.24.so > >> 0x7fdf5d5ae000 34K > >/lib/x86_64-linux-gnu/libnss_compat-2.24.so > >> 0x7fdf5d7b7000 187K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libjava.so > >> 0x7fdf5d9e6000 70K > >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libverify.so > >> 0x000
Re: Solr uses lots of shared memory!
Hi all Malloc has a lock while it is active in the heap. If there is more than one thread, and malloc finds the lock in use, then it avoids waiting on the lock by creating a new 'arena' to hold its heap. My understanding is that a process with multiple threads which are all active users of malloc will eventually have an arena per thread. If you limit the number of arenas, you may suffer delays waiting on locks. But this needs performance testing. My experience was with C++, not with a JVM. I would be interested to know if setting MALLOC_ARENA_MAX=2 makes a difference to performance. Cheers -- Rick On September 2, 2017 1:15:38 PM EDT, Kevin Risden wrote: >I haven't looked at reproducing this locally, but since it seems like >there haven't been any new ideas decided to share this in case it >helps: > >I noticed in Travis CI [1] they are adding the environment variable >MALLOC_ARENA_MAX=2 and so I googled what that configuration did. To my >surprise, I came across a stackoverflow post [2] about how glibc could >actually be the case and report memory differently. I then found a >Hadoop issue HADOOP-7154 [3] about setting this as well to reduce >virtual memory usage. I found some more cases where this has helped as >well [4], [5], and [6] > >[1] >https://docs.travis-ci.com/user/build-environment-updates/2017-09-06/#Added >[2] >https://stackoverflow.com/questions/10575342/what-would-cause-a-java-process-to-greatly-exceed-the-xmx-or-xss-limit >[3] >https://issues.apache.org/jira/browse/HADOOP-7154?focusedCommentId=14505792 >[4] https://github.com/cloudfoundry/java-buildpack/issues/320 >[5] https://devcenter.heroku.com/articles/tuning-glibc-memory-behavior >[6] >https://www.ibm.com/developerworks/community/blogs/kevgrig/entry/linux_glibc_2_10_rhel_6_malloc_may_show_excessive_virtual_memory_usage?lang=en >Kevin Risden > > >On Thu, Aug 24, 2017 at 10:19 AM, Markus Jelsma > wrote: >> Hello Bernd, >> >> According to the man page, i should get a list of stuff in shared >memory if i invoke it with just a PID. Which shows a list of libraries >that together account for about 25 MB's shared memory usage. Accoring >to ps and top, the JVM uses 2800 MB shared memory (not virtual), that >leaves 2775 MB unaccounted for. Any ideas? Anyone else to reproduce it >on a freshly restarted node? >> >> Thanks, >> Markus >> >> >> PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ >COMMAND >> 18901 markus20 0 14,778g 4,965g 2,987g S 891,1 31,7 20:21.63 >java >> >> 0x55b9a17f1000 6K >/usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java >> 0x7fdf1d314000 182K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libsunec.so >> 0x7fdf1e548000 38K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libmanagement.so >> 0x7fdf1e78e000 94K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnet.so >> 0x7fdf1e9a6000 75K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnio.so >> 0x7fdf5cd6e000 34K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libzip.so >> 0x7fdf5cf77000 46K >/lib/x86_64-linux-gnu/libnss_files-2.24.so >> 0x7fdf5d189000 46K >/lib/x86_64-linux-gnu/libnss_nis-2.24.so >> 0x7fdf5d395000 90K /lib/x86_64-linux-gnu/libnsl-2.24.so >> 0x7fdf5d5ae000 34K >/lib/x86_64-linux-gnu/libnss_compat-2.24.so >> 0x7fdf5d7b7000 187K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libjava.so >> 0x7fdf5d9e6000 70K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libverify.so >> 0x7fdf5dbf8000 30K /lib/x86_64-linux-gnu/librt-2.24.so >> 0x7fdf5de0 90K /lib/x86_64-linux-gnu/libgcc_s.so.1 >> 0x7fdf5e017000 1063K /lib/x86_64-linux-gnu/libm-2.24.so >> 0x7fdf5e32 1553K >/usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.22 >> 0x7fdf5e6a8000 15936K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/server/libjvm.so >> 0x7fdf5f5ed000 139K >/lib/x86_64-linux-gnu/libpthread-2.24.so >> 0x7fdf5f80b000 14K /lib/x86_64-linux-gnu/libdl-2.24.so >> 0x7fdf5fa0f000 110K/lib/x86_64-linux-gnu/libz.so.1.2.11 >> 0x7fdf5fc2b000 1813K /lib/x86_64-linux-gnu/libc-2.24.so >> 0x7fdf5fff2000 58K >/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/jli/libjli.so >> 0x7fdf60201000 158K/lib/x86_64-linux-gnu/ld-2.24.so >> >> -Original message- >>> From:Bernd Fehling >>> Sent: Thursday 24th August 2017 15:39 >>> To: solr-user@lucene.apache.org >>> Subjec
Re: Solr uses lots of shared memory!
I haven't looked at reproducing this locally, but since it seems like there haven't been any new ideas decided to share this in case it helps: I noticed in Travis CI [1] they are adding the environment variable MALLOC_ARENA_MAX=2 and so I googled what that configuration did. To my surprise, I came across a stackoverflow post [2] about how glibc could actually be the case and report memory differently. I then found a Hadoop issue HADOOP-7154 [3] about setting this as well to reduce virtual memory usage. I found some more cases where this has helped as well [4], [5], and [6] [1] https://docs.travis-ci.com/user/build-environment-updates/2017-09-06/#Added [2] https://stackoverflow.com/questions/10575342/what-would-cause-a-java-process-to-greatly-exceed-the-xmx-or-xss-limit [3] https://issues.apache.org/jira/browse/HADOOP-7154?focusedCommentId=14505792 [4] https://github.com/cloudfoundry/java-buildpack/issues/320 [5] https://devcenter.heroku.com/articles/tuning-glibc-memory-behavior [6] https://www.ibm.com/developerworks/community/blogs/kevgrig/entry/linux_glibc_2_10_rhel_6_malloc_may_show_excessive_virtual_memory_usage?lang=en Kevin Risden On Thu, Aug 24, 2017 at 10:19 AM, Markus Jelsma wrote: > Hello Bernd, > > According to the man page, i should get a list of stuff in shared memory if i > invoke it with just a PID. Which shows a list of libraries that together > account for about 25 MB's shared memory usage. Accoring to ps and top, the > JVM uses 2800 MB shared memory (not virtual), that leaves 2775 MB unaccounted > for. Any ideas? Anyone else to reproduce it on a freshly restarted node? > > Thanks, > Markus > > > PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND > 18901 markus20 0 14,778g 4,965g 2,987g S 891,1 31,7 20:21.63 java > > 0x55b9a17f1000 6K /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java > 0x7fdf1d314000 182K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libsunec.so > 0x7fdf1e548000 38K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libmanagement.so > 0x7fdf1e78e000 94K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnet.so > 0x7fdf1e9a6000 75K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnio.so > 0x7fdf5cd6e000 34K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libzip.so > 0x7fdf5cf77000 46K /lib/x86_64-linux-gnu/libnss_files-2.24.so > 0x7fdf5d189000 46K /lib/x86_64-linux-gnu/libnss_nis-2.24.so > 0x7fdf5d395000 90K /lib/x86_64-linux-gnu/libnsl-2.24.so > 0x7fdf5d5ae000 34K /lib/x86_64-linux-gnu/libnss_compat-2.24.so > 0x7fdf5d7b7000 187K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libjava.so > 0x7fdf5d9e6000 70K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libverify.so > 0x7fdf5dbf8000 30K /lib/x86_64-linux-gnu/librt-2.24.so > 0x7fdf5de0 90K /lib/x86_64-linux-gnu/libgcc_s.so.1 > 0x7fdf5e017000 1063K /lib/x86_64-linux-gnu/libm-2.24.so > 0x7fdf5e32 1553K /usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.22 > 0x7fdf5e6a8000 15936K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/server/libjvm.so > 0x7fdf5f5ed000 139K/lib/x86_64-linux-gnu/libpthread-2.24.so > 0x7fdf5f80b000 14K /lib/x86_64-linux-gnu/libdl-2.24.so > 0x7fdf5fa0f000 110K/lib/x86_64-linux-gnu/libz.so.1.2.11 > 0x7fdf5fc2b000 1813K /lib/x86_64-linux-gnu/libc-2.24.so > 0x7fdf5fff2000 58K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/jli/libjli.so > 0x7fdf60201000 158K/lib/x86_64-linux-gnu/ld-2.24.so > > -Original message- >> From:Bernd Fehling >> Sent: Thursday 24th August 2017 15:39 >> To: solr-user@lucene.apache.org >> Subject: Re: Solr uses lots of shared memory! >> >> Just an idea, how about taking a dump with jmap and using >> MemoryAnalyzerTool to see what is going on? >> >> Regards >> Bernd >> >> >> Am 24.08.2017 um 11:49 schrieb Markus Jelsma: >> > Hello Shalin, >> > >> > Yes, the main search index has DocValues on just a few fields, they are >> > used for facetting and function queries, we started using DocValues when >> > 6.0 was released. Most fields are content fields for many languages. I >> > don't think it is going to be DocValues because the max shared memory >> > consumption is reduced my searching on fields fewer languages, and by >> > disabling highlighting, both not using DocValues. >> > >> > But it tried the option regardless, and because i didn't know about it. >> > But it appears the op
RE: Solr uses lots of shared memory!
Anyone else besides Shawn and me to reproduce this problem? Shawn contacted Oracle off-list but that was useless at best (attach JConsole, watch heap, etc). Is this a real problem, just a bad reporting issue of the JVM and Linux? Thanks, Markus -Original message- > From:Markus Jelsma > Sent: Thursday 24th August 2017 17:20 > To: solr-user@lucene.apache.org > Subject: RE: Solr uses lots of shared memory! > > Hello Bernd, > > According to the man page, i should get a list of stuff in shared memory if i > invoke it with just a PID. Which shows a list of libraries that together > account for about 25 MB's shared memory usage. Accoring to ps and top, the > JVM uses 2800 MB shared memory (not virtual), that leaves 2775 MB unaccounted > for. Any ideas? Anyone else to reproduce it on a freshly restarted node? > > Thanks, > Markus > > > PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND > > > 18901 markus20 0 14,778g 4,965g 2,987g S 891,1 31,7 20:21.63 java > > 0x55b9a17f1000 6K /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java > 0x7fdf1d314000 182K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libsunec.so > 0x7fdf1e548000 38K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libmanagement.so > 0x7fdf1e78e000 94K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnet.so > 0x7fdf1e9a6000 75K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnio.so > 0x7fdf5cd6e000 34K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libzip.so > 0x7fdf5cf77000 46K /lib/x86_64-linux-gnu/libnss_files-2.24.so > 0x7fdf5d189000 46K /lib/x86_64-linux-gnu/libnss_nis-2.24.so > 0x7fdf5d395000 90K /lib/x86_64-linux-gnu/libnsl-2.24.so > 0x7fdf5d5ae000 34K /lib/x86_64-linux-gnu/libnss_compat-2.24.so > 0x7fdf5d7b7000 187K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libjava.so > 0x7fdf5d9e6000 70K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libverify.so > 0x7fdf5dbf8000 30K /lib/x86_64-linux-gnu/librt-2.24.so > 0x7fdf5de0 90K /lib/x86_64-linux-gnu/libgcc_s.so.1 > 0x7fdf5e017000 1063K /lib/x86_64-linux-gnu/libm-2.24.so > 0x7fdf5e32 1553K /usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.22 > 0x7fdf5e6a8000 15936K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/server/libjvm.so > 0x7fdf5f5ed000 139K/lib/x86_64-linux-gnu/libpthread-2.24.so > 0x7fdf5f80b000 14K /lib/x86_64-linux-gnu/libdl-2.24.so > 0x7fdf5fa0f000 110K/lib/x86_64-linux-gnu/libz.so.1.2.11 > 0x7fdf5fc2b000 1813K /lib/x86_64-linux-gnu/libc-2.24.so > 0x7fdf5fff2000 58K > /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/jli/libjli.so > 0x7fdf60201000 158K/lib/x86_64-linux-gnu/ld-2.24.so > > -Original message- > > From:Bernd Fehling > > Sent: Thursday 24th August 2017 15:39 > > To: solr-user@lucene.apache.org > > Subject: Re: Solr uses lots of shared memory! > > > > Just an idea, how about taking a dump with jmap and using > > MemoryAnalyzerTool to see what is going on? > > > > Regards > > Bernd > > > > > > Am 24.08.2017 um 11:49 schrieb Markus Jelsma: > > > Hello Shalin, > > > > > > Yes, the main search index has DocValues on just a few fields, they are > > > used for facetting and function queries, we started using DocValues when > > > 6.0 was released. Most fields are content fields for many languages. I > > > don't think it is going to be DocValues because the max shared memory > > > consumption is reduced my searching on fields fewer languages, and by > > > disabling highlighting, both not using DocValues. > > > > > > But it tried the option regardless, and because i didn't know about it. > > > But it appears the option does exactly nothing. First is without any > > > configuration for preload, second is with preload=true, third is > > > preload=false > > > > > > 14220 markus20 0 14,675g 1,508g 62800 S 1,0 9,6 0:36.98 java > > > 14803 markus20 0 14,674g 1,537g 63248 S 0,0 9,8 0:34.50 java > > > 15324 markus20 0 14,674g 1,409g 63152 S 0,0 9,0 0:35.50 java > > > > > > Please correct my config is i am wrong: > > > > > >> > class="${solr.directoryFactory:solr.NRTCachingDirectoryFactory
RE: Solr uses lots of shared memory!
Hello Bernd, According to the man page, i should get a list of stuff in shared memory if i invoke it with just a PID. Which shows a list of libraries that together account for about 25 MB's shared memory usage. Accoring to ps and top, the JVM uses 2800 MB shared memory (not virtual), that leaves 2775 MB unaccounted for. Any ideas? Anyone else to reproduce it on a freshly restarted node? Thanks, Markus PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND 18901 markus20 0 14,778g 4,965g 2,987g S 891,1 31,7 20:21.63 java 0x55b9a17f1000 6K /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java 0x7fdf1d314000 182K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libsunec.so 0x7fdf1e548000 38K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libmanagement.so 0x7fdf1e78e000 94K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnet.so 0x7fdf1e9a6000 75K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libnio.so 0x7fdf5cd6e000 34K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libzip.so 0x7fdf5cf77000 46K /lib/x86_64-linux-gnu/libnss_files-2.24.so 0x7fdf5d189000 46K /lib/x86_64-linux-gnu/libnss_nis-2.24.so 0x7fdf5d395000 90K /lib/x86_64-linux-gnu/libnsl-2.24.so 0x7fdf5d5ae000 34K /lib/x86_64-linux-gnu/libnss_compat-2.24.so 0x7fdf5d7b7000 187K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libjava.so 0x7fdf5d9e6000 70K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libverify.so 0x7fdf5dbf8000 30K /lib/x86_64-linux-gnu/librt-2.24.so 0x7fdf5de0 90K /lib/x86_64-linux-gnu/libgcc_s.so.1 0x7fdf5e017000 1063K /lib/x86_64-linux-gnu/libm-2.24.so 0x7fdf5e32 1553K /usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.22 0x7fdf5e6a8000 15936K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/server/libjvm.so 0x7fdf5f5ed000 139K/lib/x86_64-linux-gnu/libpthread-2.24.so 0x7fdf5f80b000 14K /lib/x86_64-linux-gnu/libdl-2.24.so 0x7fdf5fa0f000 110K/lib/x86_64-linux-gnu/libz.so.1.2.11 0x7fdf5fc2b000 1813K /lib/x86_64-linux-gnu/libc-2.24.so 0x7fdf5fff2000 58K /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/jli/libjli.so 0x7fdf60201000 158K/lib/x86_64-linux-gnu/ld-2.24.so -Original message- > From:Bernd Fehling > Sent: Thursday 24th August 2017 15:39 > To: solr-user@lucene.apache.org > Subject: Re: Solr uses lots of shared memory! > > Just an idea, how about taking a dump with jmap and using > MemoryAnalyzerTool to see what is going on? > > Regards > Bernd > > > Am 24.08.2017 um 11:49 schrieb Markus Jelsma: > > Hello Shalin, > > > > Yes, the main search index has DocValues on just a few fields, they are > > used for facetting and function queries, we started using DocValues when > > 6.0 was released. Most fields are content fields for many languages. I > > don't think it is going to be DocValues because the max shared memory > > consumption is reduced my searching on fields fewer languages, and by > > disabling highlighting, both not using DocValues. > > > > But it tried the option regardless, and because i didn't know about it. But > > it appears the option does exactly nothing. First is without any > > configuration for preload, second is with preload=true, third is > > preload=false > > > > 14220 markus20 0 14,675g 1,508g 62800 S 1,0 9,6 0:36.98 java > > 14803 markus20 0 14,674g 1,537g 63248 S 0,0 9,8 0:34.50 java > > 15324 markus20 0 14,674g 1,409g 63152 S 0,0 9,0 0:35.50 java > > > > Please correct my config is i am wrong: > > > >> class="${solr.directoryFactory:solr.NRTCachingDirectoryFactory}"> > > false > > > > > > NRTCachingDirectoryFactory implies MMapDirectory right? > > > > Thanks, > > Markus > > > > -Original message- > >> From:Shalin Shekhar Mangar > >> Sent: Thursday 24th August 2017 5:51 > >> To: solr-user@lucene.apache.org > >> Subject: Re: Solr uses lots of shared memory! > >> > >> Very interesting. Do you have many DocValue fields? Have you always > >> had them i.e. did you see this problem before you turned on DocValues? > >> The DocValue fields are in a separate file and they will be memory > >> mapped on demand. One thing you can experiment with is to use > >> preload=true option on the MMapDirectoryFactory which will mmap all > >> index files on startu
Re: Solr uses lots of shared memory!
Just an idea, how about taking a dump with jmap and using MemoryAnalyzerTool to see what is going on? Regards Bernd Am 24.08.2017 um 11:49 schrieb Markus Jelsma: > Hello Shalin, > > Yes, the main search index has DocValues on just a few fields, they are used > for facetting and function queries, we started using DocValues when 6.0 was > released. Most fields are content fields for many languages. I don't think it > is going to be DocValues because the max shared memory consumption is reduced > my searching on fields fewer languages, and by disabling highlighting, both > not using DocValues. > > But it tried the option regardless, and because i didn't know about it. But > it appears the option does exactly nothing. First is without any > configuration for preload, second is with preload=true, third is preload=false > > 14220 markus20 0 14,675g 1,508g 62800 S 1,0 9,6 0:36.98 java > 14803 markus20 0 14,674g 1,537g 63248 S 0,0 9,8 0:34.50 java > 15324 markus20 0 14,674g 1,409g 63152 S 0,0 9,0 0:35.50 java > > Please correct my config is i am wrong: > >class="${solr.directoryFactory:solr.NRTCachingDirectoryFactory}"> > false > > > NRTCachingDirectoryFactory implies MMapDirectory right? > > Thanks, > Markus > > -Original message- >> From:Shalin Shekhar Mangar >> Sent: Thursday 24th August 2017 5:51 >> To: solr-user@lucene.apache.org >> Subject: Re: Solr uses lots of shared memory! >> >> Very interesting. Do you have many DocValue fields? Have you always >> had them i.e. did you see this problem before you turned on DocValues? >> The DocValue fields are in a separate file and they will be memory >> mapped on demand. One thing you can experiment with is to use >> preload=true option on the MMapDirectoryFactory which will mmap all >> index files on startup [1]. Once you do this, and if you still notice >> shared memory leakage then it may be a genuine memory leak that we >> should investigate. >> >> [1] - >> http://lucene.apache.org/solr/guide/6_6/datadir-and-directoryfactory-in-solrconfig.html#DataDirandDirectoryFactoryinSolrConfig-SpecifyingtheDirectoryFactoryForYourIndex >> >> On Wed, Aug 23, 2017 at 7:02 PM, Markus Jelsma >> wrote: >>> I do not think it is a problem of reporting after watching top after >>> restart of some Solr instances, it dropped back to `normal`, around 350 MB, >>> which i think it high to but anyway. >>> >>> Two hours later, the restarted nodes are slowly increasing shared memory >>> consumption to about 1500 MB now. I don't understand why shared memory >>> usage should/would increase slowly over time, it makes little sense to me >>> and i cannot remember Solr doing this in the past ten years. >>> >>> But it seems to correlate to index size on disk, these main text search >>> nodes have an index of around 16 GB and up 3 GB of shared memory after a >>> few days. Logs nodes up to 800 MB index size and 320 MB of shared memory, >>> the low latency nodes have four different cores that make up just over 100 >>> MB index size, shared memory consumption is just 22 MB, which seems more >>> reasonable for the case of shared memory. >>> >>> I can also force Solr to 'leak' shared memory just by sending queries to >>> it. My freshly restarted local node used 68 MB shared memory at startup. >>> Two minutes and 25.000 queries later it was already 2748 MB! At first there >>> is a very sharp increase to 2000, then it takes almost two minutes more to >>> increase to 2748. I can decrease the maximum shared memory usage to 1200 if >>> i query (via edismax) only on fields of one language instead of 25 orso. I >>> can decrease it as well further if i disable highlighting (HUH?) but still >>> query on all fields. >>> >>> * We have tried patching Java's ByteBuffer [1] because it seemed to fit the >>> problems, it does not fix it. >>> * We have also removed all our custom plugins, so it has become a vanilla >>> Solr 6.6 just with our stripped down schema and solrconfig, it neither >>> fixes it. >>> >>> Why does it slowly increase over time? >>> Why does it appear to correlate to index size? >>> Is anyone else seeing this on their 6.6 cloud production or local machines? >>> >>> Thanks, >>> Markus >>> >>> [1]: http://www.evanjones.ca/java-bytebuffer-leak.html >>> >>> -Original message- >>>> Fro
RE: Solr uses lots of shared memory!
Hello Shalin, Yes, the main search index has DocValues on just a few fields, they are used for facetting and function queries, we started using DocValues when 6.0 was released. Most fields are content fields for many languages. I don't think it is going to be DocValues because the max shared memory consumption is reduced my searching on fields fewer languages, and by disabling highlighting, both not using DocValues. But it tried the option regardless, and because i didn't know about it. But it appears the option does exactly nothing. First is without any configuration for preload, second is with preload=true, third is preload=false 14220 markus20 0 14,675g 1,508g 62800 S 1,0 9,6 0:36.98 java 14803 markus20 0 14,674g 1,537g 63248 S 0,0 9,8 0:34.50 java 15324 markus20 0 14,674g 1,409g 63152 S 0,0 9,0 0:35.50 java Please correct my config is i am wrong: false NRTCachingDirectoryFactory implies MMapDirectory right? Thanks, Markus -Original message- > From:Shalin Shekhar Mangar > Sent: Thursday 24th August 2017 5:51 > To: solr-user@lucene.apache.org > Subject: Re: Solr uses lots of shared memory! > > Very interesting. Do you have many DocValue fields? Have you always > had them i.e. did you see this problem before you turned on DocValues? > The DocValue fields are in a separate file and they will be memory > mapped on demand. One thing you can experiment with is to use > preload=true option on the MMapDirectoryFactory which will mmap all > index files on startup [1]. Once you do this, and if you still notice > shared memory leakage then it may be a genuine memory leak that we > should investigate. > > [1] - > http://lucene.apache.org/solr/guide/6_6/datadir-and-directoryfactory-in-solrconfig.html#DataDirandDirectoryFactoryinSolrConfig-SpecifyingtheDirectoryFactoryForYourIndex > > On Wed, Aug 23, 2017 at 7:02 PM, Markus Jelsma > wrote: > > I do not think it is a problem of reporting after watching top after > > restart of some Solr instances, it dropped back to `normal`, around 350 MB, > > which i think it high to but anyway. > > > > Two hours later, the restarted nodes are slowly increasing shared memory > > consumption to about 1500 MB now. I don't understand why shared memory > > usage should/would increase slowly over time, it makes little sense to me > > and i cannot remember Solr doing this in the past ten years. > > > > But it seems to correlate to index size on disk, these main text search > > nodes have an index of around 16 GB and up 3 GB of shared memory after a > > few days. Logs nodes up to 800 MB index size and 320 MB of shared memory, > > the low latency nodes have four different cores that make up just over 100 > > MB index size, shared memory consumption is just 22 MB, which seems more > > reasonable for the case of shared memory. > > > > I can also force Solr to 'leak' shared memory just by sending queries to > > it. My freshly restarted local node used 68 MB shared memory at startup. > > Two minutes and 25.000 queries later it was already 2748 MB! At first there > > is a very sharp increase to 2000, then it takes almost two minutes more to > > increase to 2748. I can decrease the maximum shared memory usage to 1200 if > > i query (via edismax) only on fields of one language instead of 25 orso. I > > can decrease it as well further if i disable highlighting (HUH?) but still > > query on all fields. > > > > * We have tried patching Java's ByteBuffer [1] because it seemed to fit the > > problems, it does not fix it. > > * We have also removed all our custom plugins, so it has become a vanilla > > Solr 6.6 just with our stripped down schema and solrconfig, it neither > > fixes it. > > > > Why does it slowly increase over time? > > Why does it appear to correlate to index size? > > Is anyone else seeing this on their 6.6 cloud production or local machines? > > > > Thanks, > > Markus > > > > [1]: http://www.evanjones.ca/java-bytebuffer-leak.html > > > > -Original message- > >> From:Shawn Heisey > >> Sent: Tuesday 22nd August 2017 17:32 > >> To: solr-user@lucene.apache.org > >> Subject: Re: Solr uses lots of shared memory! > >> > >> On 8/22/2017 7:24 AM, Markus Jelsma wrote: > >> > I have never seen this before, one of our collections, all nodes eating > >> > tons of shared memory! > >> > > >> > Here's one of the nodes: > >> > 10497 solr 20 0 19.439g 4.505g 3.139g S 1.0 57.8 2511:46 java > >> > > >> > RSS is roughly
RE: Solr uses lots of shared memory!
Hello Erick, I know the article, it is about virtual memory. My problem is with shared memory. Correct me if i am wrong, but MMApped files do not occupy shared but virtual instead. If i am wrong, the article must be rewritten. Our main searchers show very normal numbers for virtual, which is index size plus RSS. My problem is that RSS is more than four times our heap size. Usually RSS is heap + PermGen/metaspace + code cache + (threads * stack size) + some other stuff, so a 1 GB heap would consume about 1500 MB tops. Our RSS for Solr is 4.3 GB of which 2.85 GB is shared memory. Let me know what you think. Thanks, Markus -Original message- > From:Erick Erickson > Sent: Thursday 24th August 2017 3:35 > To: solr-user > Subject: Re: Solr uses lots of shared memory! > > I suspect you've already seen this, but top and similar can be > confusing when trying to interpret MMapDirectory. Uwe has an excellent > explication: > > http://blog.thetaphi.de/2012/07/use-lucenes-mmapdirectory-on-64bit.html > > Best, > Erick > > On Wed, Aug 23, 2017 at 9:10 AM, Markus Jelsma > wrote: > > I have the problem in production and local, with default Solr 6.6 JVM > > arguments, environments are: > > > > openjdk version "1.8.0_141" > > OpenJDK Runtime Environment (build 1.8.0_141-8u141-b15-1~deb9u1-b15) > > OpenJDK 64-Bit Server VM (build 25.141-b15, mixed mode) > > Linux idx1 4.9.0-3-amd64 #1 SMP Debian 4.9.30-2+deb9u1 (2017-06-18) x86_64 > > GNU/Linux > > > > and > > > > openjdk version "1.8.0_131" > > OpenJDK Runtime Environment (build 1.8.0_131-8u131-b11-2ubuntu1.17.04.3-b11) > > OpenJDK 64-Bit Server VM (build 25.131-b11, mixed mode) > > Linux midas 4.10.0-32-generic #36-Ubuntu SMP Tue Aug 8 12:10:06 UTC 2017 > > x86_64 x86_64 x86_64 GNU/Linux > > > > Regarding the node that shows the problem, can you reproduce it locally? > > Fire it up, put some data in, confirm low shared space usage, and execute > > few thousands queries against it? We immediately see a sharp rise in shared > > memory, MB's per second until it reaches some sort of plateau. > > > > -Original message- > >> From:Shawn Heisey > >> Sent: Wednesday 23rd August 2017 16:37 > >> To: solr-user@lucene.apache.org > >> Subject: Re: Solr uses lots of shared memory! > >> > >> On 8/23/2017 7:32 AM, Markus Jelsma wrote: > >> > Why does it slowly increase over time? > >> > Why does it appear to correlate to index size? > >> > Is anyone else seeing this on their 6.6 cloud production or local > >> > machines? > >> > >> More detailed information included here. My 6.6 dev install is NOT > >> having the problem, but a much older version IS. > >> > >> I grabbed this screenshot only moments ago from a production server > >> which is exhibiting a large SHR value for the Solr process: > >> > >> https://www.dropbox.com/s/q79lo2gft9es06u/idxa1-top-big-shr.png?dl=0 > >> > >> This is Solr 4.7.2, with a 10 month uptime for the Solr process, running > >> with these arguments: > >> > >> -DSTOP.KEY=REDACTED > >> -DSTOP.PORT=8078 > >> -Djetty.port=8981 > >> -Dsolr.solr.home=/index/solr4 > >> -Dcom.sun.management.jmxremote.authenticate=false > >> -Dcom.sun.management.jmxremote.ssl=false > >> -Dcom.sun.management.jmxremote.port=8686 > >> -Dcom.sun.management.jmxremote > >> -XX:+PrintReferenceGC > >> -XX:+PrintAdaptiveSizePolicy > >> -XX:+PrintGCDetails > >> -XX:+PrintGCDateStamps > >> -Xloggc:logs/gc.log-verbose:gc > >> -XX:+AggressiveOpts > >> -XX:+UseLargePages > >> -XX:InitiatingHeapOccupancyPercent=75 > >> -XX:MaxGCPauseMillis=250 > >> -XX:G1HeapRegionSize=8m > >> -XX:+ParallelRefProcEnabled > >> -XX:+PerfDisableSharedMem > >> -XX:+UseG1GC > >> -Dlog4j.configuration=file:etc/log4j.properties > >> -Xmx8192M > >> -Xms4096M > >> > >> The OS is CentOS 6, with the following Java and kernel: > >> > >> java version "1.7.0_72" > >> Java(TM) SE Runtime Environment (build 1.7.0_72-b14) > >> Java HotSpot(TM) 64-Bit Server VM (build 24.72-b04, mixed mode) > >> > >> Linux idxa1 2.6.32-431.11.2.el6.centos.plus.x86_64 #1 SMP Tue Mar 25 > >> 21:36:54 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux > >> > >> I also just grabbed a screenshot from my dev server, running Ubuntu 14, > >> Solr 6.6.0, a LOT more index
Re: Solr uses lots of shared memory!
Very interesting. Do you have many DocValue fields? Have you always had them i.e. did you see this problem before you turned on DocValues? The DocValue fields are in a separate file and they will be memory mapped on demand. One thing you can experiment with is to use preload=true option on the MMapDirectoryFactory which will mmap all index files on startup [1]. Once you do this, and if you still notice shared memory leakage then it may be a genuine memory leak that we should investigate. [1] - http://lucene.apache.org/solr/guide/6_6/datadir-and-directoryfactory-in-solrconfig.html#DataDirandDirectoryFactoryinSolrConfig-SpecifyingtheDirectoryFactoryForYourIndex On Wed, Aug 23, 2017 at 7:02 PM, Markus Jelsma wrote: > I do not think it is a problem of reporting after watching top after restart > of some Solr instances, it dropped back to `normal`, around 350 MB, which i > think it high to but anyway. > > Two hours later, the restarted nodes are slowly increasing shared memory > consumption to about 1500 MB now. I don't understand why shared memory usage > should/would increase slowly over time, it makes little sense to me and i > cannot remember Solr doing this in the past ten years. > > But it seems to correlate to index size on disk, these main text search nodes > have an index of around 16 GB and up 3 GB of shared memory after a few days. > Logs nodes up to 800 MB index size and 320 MB of shared memory, the low > latency nodes have four different cores that make up just over 100 MB index > size, shared memory consumption is just 22 MB, which seems more reasonable > for the case of shared memory. > > I can also force Solr to 'leak' shared memory just by sending queries to it. > My freshly restarted local node used 68 MB shared memory at startup. Two > minutes and 25.000 queries later it was already 2748 MB! At first there is a > very sharp increase to 2000, then it takes almost two minutes more to > increase to 2748. I can decrease the maximum shared memory usage to 1200 if i > query (via edismax) only on fields of one language instead of 25 orso. I can > decrease it as well further if i disable highlighting (HUH?) but still query > on all fields. > > * We have tried patching Java's ByteBuffer [1] because it seemed to fit the > problems, it does not fix it. > * We have also removed all our custom plugins, so it has become a vanilla > Solr 6.6 just with our stripped down schema and solrconfig, it neither fixes > it. > > Why does it slowly increase over time? > Why does it appear to correlate to index size? > Is anyone else seeing this on their 6.6 cloud production or local machines? > > Thanks, > Markus > > [1]: http://www.evanjones.ca/java-bytebuffer-leak.html > > -----Original message- >> From:Shawn Heisey >> Sent: Tuesday 22nd August 2017 17:32 >> To: solr-user@lucene.apache.org >> Subject: Re: Solr uses lots of shared memory! >> >> On 8/22/2017 7:24 AM, Markus Jelsma wrote: >> > I have never seen this before, one of our collections, all nodes eating >> > tons of shared memory! >> > >> > Here's one of the nodes: >> > 10497 solr 20 0 19.439g 4.505g 3.139g S 1.0 57.8 2511:46 java >> > >> > RSS is roughly equal to heap size + usual off-heap space + shared memory. >> > Virtual is equal to RSS and index size on disk. For two other collections, >> > the nodes use shared memory as expected, in the MB range. >> > >> > How can Solr, this collection, use so much shared memory? Why? >> >> I've seen this on my own servers at work, and when I add up a subset of >> the memory numbers I can see from the system, it ends up being more >> memory than I even have in the server. >> >> I suspect there is something odd going on in how Java reports memory >> usage to the OS, or maybe a glitch in how Linux interprets Java's memory >> usage. At some point in the past, numbers were reported correctly. I >> do not know if the change came about because of a Solr upgrade, because >> of a Java upgrade, or because of an OS kernel upgrade. All three were >> upgraded between when I know the numbers looked right and when I noticed >> they were wrong. >> >> https://www.dropbox.com/s/91uqlrnfghr2heo/solr-memory-sorted-top.png?dl=0 >> >> This screenshot shows that Solr is using 17GB of memory, 41.45GB of >> memory is being used by the OS disk cache, and 10.23GB of memory is >> free. Add those up, and it comes to 68.68GB ... but the machine only >> has 64GB of memory, and that total doesn't include the memory usage of >> the other processes seen in the screenshot. This impossible
Re: Solr uses lots of shared memory!
I suspect you've already seen this, but top and similar can be confusing when trying to interpret MMapDirectory. Uwe has an excellent explication: http://blog.thetaphi.de/2012/07/use-lucenes-mmapdirectory-on-64bit.html Best, Erick On Wed, Aug 23, 2017 at 9:10 AM, Markus Jelsma wrote: > I have the problem in production and local, with default Solr 6.6 JVM > arguments, environments are: > > openjdk version "1.8.0_141" > OpenJDK Runtime Environment (build 1.8.0_141-8u141-b15-1~deb9u1-b15) > OpenJDK 64-Bit Server VM (build 25.141-b15, mixed mode) > Linux idx1 4.9.0-3-amd64 #1 SMP Debian 4.9.30-2+deb9u1 (2017-06-18) x86_64 > GNU/Linux > > and > > openjdk version "1.8.0_131" > OpenJDK Runtime Environment (build 1.8.0_131-8u131-b11-2ubuntu1.17.04.3-b11) > OpenJDK 64-Bit Server VM (build 25.131-b11, mixed mode) > Linux midas 4.10.0-32-generic #36-Ubuntu SMP Tue Aug 8 12:10:06 UTC 2017 > x86_64 x86_64 x86_64 GNU/Linux > > Regarding the node that shows the problem, can you reproduce it locally? Fire > it up, put some data in, confirm low shared space usage, and execute few > thousands queries against it? We immediately see a sharp rise in shared > memory, MB's per second until it reaches some sort of plateau. > > -Original message- >> From:Shawn Heisey >> Sent: Wednesday 23rd August 2017 16:37 >> To: solr-user@lucene.apache.org >> Subject: Re: Solr uses lots of shared memory! >> >> On 8/23/2017 7:32 AM, Markus Jelsma wrote: >> > Why does it slowly increase over time? >> > Why does it appear to correlate to index size? >> > Is anyone else seeing this on their 6.6 cloud production or local machines? >> >> More detailed information included here. My 6.6 dev install is NOT >> having the problem, but a much older version IS. >> >> I grabbed this screenshot only moments ago from a production server >> which is exhibiting a large SHR value for the Solr process: >> >> https://www.dropbox.com/s/q79lo2gft9es06u/idxa1-top-big-shr.png?dl=0 >> >> This is Solr 4.7.2, with a 10 month uptime for the Solr process, running >> with these arguments: >> >> -DSTOP.KEY=REDACTED >> -DSTOP.PORT=8078 >> -Djetty.port=8981 >> -Dsolr.solr.home=/index/solr4 >> -Dcom.sun.management.jmxremote.authenticate=false >> -Dcom.sun.management.jmxremote.ssl=false >> -Dcom.sun.management.jmxremote.port=8686 >> -Dcom.sun.management.jmxremote >> -XX:+PrintReferenceGC >> -XX:+PrintAdaptiveSizePolicy >> -XX:+PrintGCDetails >> -XX:+PrintGCDateStamps >> -Xloggc:logs/gc.log-verbose:gc >> -XX:+AggressiveOpts >> -XX:+UseLargePages >> -XX:InitiatingHeapOccupancyPercent=75 >> -XX:MaxGCPauseMillis=250 >> -XX:G1HeapRegionSize=8m >> -XX:+ParallelRefProcEnabled >> -XX:+PerfDisableSharedMem >> -XX:+UseG1GC >> -Dlog4j.configuration=file:etc/log4j.properties >> -Xmx8192M >> -Xms4096M >> >> The OS is CentOS 6, with the following Java and kernel: >> >> java version "1.7.0_72" >> Java(TM) SE Runtime Environment (build 1.7.0_72-b14) >> Java HotSpot(TM) 64-Bit Server VM (build 24.72-b04, mixed mode) >> >> Linux idxa1 2.6.32-431.11.2.el6.centos.plus.x86_64 #1 SMP Tue Mar 25 >> 21:36:54 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux >> >> I also just grabbed a screenshot from my dev server, running Ubuntu 14, >> Solr 6.6.0, a LOT more index data, and a more recent Java version. Solr >> has an uptime of about one month. This server was installed with the >> service installer script, so it uses bin/solr. It doesn't seem to have >> the same problem: >> >> https://www.dropbox.com/s/85h1weuopa643za/bigindy5-top-small-shr.png?dl=0 >> >> java version "1.8.0_144" >> Java(TM) SE Runtime Environment (build 1.8.0_144-b01) >> Java HotSpot(TM) 64-Bit Server VM (build 25.144-b01, mixed mode) >> >> Linux bigindy5 3.13.0-105-generic #152-Ubuntu SMP Fri Dec 2 15:37:11 UTC >> 2016 x86_64 x86_64 x86_64 GNU/Linux >> >> The arguments for this one are very similar to the production server: >> >> -DSTOP.KEY=solrrocks >> -DSTOP.PORT=7982 >> -Dcom.sun.management.jmxremote >> -Dcom.sun.management.jmxremote.authenticate=false >> -Dcom.sun.management.jmxremote.local.only=false >> -Dcom.sun.management.jmxremote.port=18982 >> -Dcom.sun.management.jmxremote.rmi.port=18982 >> -Dcom.sun.management.jmxremote.ssl=false >> -Djetty.home=/opt/solr6/server >> -Djetty.port=8982 >> -Dlog4j.configuration=file:/index/solr6/log4j.properties >> -Dsolr.install.dir=/opt/solr6
RE: Solr uses lots of shared memory!
I have the problem in production and local, with default Solr 6.6 JVM arguments, environments are: openjdk version "1.8.0_141" OpenJDK Runtime Environment (build 1.8.0_141-8u141-b15-1~deb9u1-b15) OpenJDK 64-Bit Server VM (build 25.141-b15, mixed mode) Linux idx1 4.9.0-3-amd64 #1 SMP Debian 4.9.30-2+deb9u1 (2017-06-18) x86_64 GNU/Linux and openjdk version "1.8.0_131" OpenJDK Runtime Environment (build 1.8.0_131-8u131-b11-2ubuntu1.17.04.3-b11) OpenJDK 64-Bit Server VM (build 25.131-b11, mixed mode) Linux midas 4.10.0-32-generic #36-Ubuntu SMP Tue Aug 8 12:10:06 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux Regarding the node that shows the problem, can you reproduce it locally? Fire it up, put some data in, confirm low shared space usage, and execute few thousands queries against it? We immediately see a sharp rise in shared memory, MB's per second until it reaches some sort of plateau. -Original message- > From:Shawn Heisey > Sent: Wednesday 23rd August 2017 16:37 > To: solr-user@lucene.apache.org > Subject: Re: Solr uses lots of shared memory! > > On 8/23/2017 7:32 AM, Markus Jelsma wrote: > > Why does it slowly increase over time? > > Why does it appear to correlate to index size? > > Is anyone else seeing this on their 6.6 cloud production or local machines? > > More detailed information included here. My 6.6 dev install is NOT > having the problem, but a much older version IS. > > I grabbed this screenshot only moments ago from a production server > which is exhibiting a large SHR value for the Solr process: > > https://www.dropbox.com/s/q79lo2gft9es06u/idxa1-top-big-shr.png?dl=0 > > This is Solr 4.7.2, with a 10 month uptime for the Solr process, running > with these arguments: > > -DSTOP.KEY=REDACTED > -DSTOP.PORT=8078 > -Djetty.port=8981 > -Dsolr.solr.home=/index/solr4 > -Dcom.sun.management.jmxremote.authenticate=false > -Dcom.sun.management.jmxremote.ssl=false > -Dcom.sun.management.jmxremote.port=8686 > -Dcom.sun.management.jmxremote > -XX:+PrintReferenceGC > -XX:+PrintAdaptiveSizePolicy > -XX:+PrintGCDetails > -XX:+PrintGCDateStamps > -Xloggc:logs/gc.log-verbose:gc > -XX:+AggressiveOpts > -XX:+UseLargePages > -XX:InitiatingHeapOccupancyPercent=75 > -XX:MaxGCPauseMillis=250 > -XX:G1HeapRegionSize=8m > -XX:+ParallelRefProcEnabled > -XX:+PerfDisableSharedMem > -XX:+UseG1GC > -Dlog4j.configuration=file:etc/log4j.properties > -Xmx8192M > -Xms4096M > > The OS is CentOS 6, with the following Java and kernel: > > java version "1.7.0_72" > Java(TM) SE Runtime Environment (build 1.7.0_72-b14) > Java HotSpot(TM) 64-Bit Server VM (build 24.72-b04, mixed mode) > > Linux idxa1 2.6.32-431.11.2.el6.centos.plus.x86_64 #1 SMP Tue Mar 25 > 21:36:54 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux > > I also just grabbed a screenshot from my dev server, running Ubuntu 14, > Solr 6.6.0, a LOT more index data, and a more recent Java version. Solr > has an uptime of about one month. This server was installed with the > service installer script, so it uses bin/solr. It doesn't seem to have > the same problem: > > https://www.dropbox.com/s/85h1weuopa643za/bigindy5-top-small-shr.png?dl=0 > > java version "1.8.0_144" > Java(TM) SE Runtime Environment (build 1.8.0_144-b01) > Java HotSpot(TM) 64-Bit Server VM (build 25.144-b01, mixed mode) > > Linux bigindy5 3.13.0-105-generic #152-Ubuntu SMP Fri Dec 2 15:37:11 UTC > 2016 x86_64 x86_64 x86_64 GNU/Linux > > The arguments for this one are very similar to the production server: > > -DSTOP.KEY=solrrocks > -DSTOP.PORT=7982 > -Dcom.sun.management.jmxremote > -Dcom.sun.management.jmxremote.authenticate=false > -Dcom.sun.management.jmxremote.local.only=false > -Dcom.sun.management.jmxremote.port=18982 > -Dcom.sun.management.jmxremote.rmi.port=18982 > -Dcom.sun.management.jmxremote.ssl=false > -Djetty.home=/opt/solr6/server > -Djetty.port=8982 > -Dlog4j.configuration=file:/index/solr6/log4j.properties > -Dsolr.install.dir=/opt/solr6 > -Dsolr.log.dir=/index/solr6/logs > -Dsolr.log.muteconsole > -Dsolr.solr.home=/index/solr6/data > -Duser.timezone=UTC > -XX:+AggressiveOpts > -XX:+ParallelRefProcEnabled > -XX:+PrintGCApplicationStoppedTime > -XX:+PrintGCDateStamps > -XX:+PrintGCDetails > -XX:+PrintGCTimeStamps > -XX:+PrintHeapAtGC > -XX:+PrintTenuringDistribution > -XX:+UseG1GC > -XX:+UseGCLogFileRotation > -XX:+UseLargePages > -XX:G1HeapRegionSize=8m > -XX:GCLogFileSize=20M > -XX:InitiatingHeapOccupancyPercent=75 > -XX:MaxGCPauseMillis=250 > -XX:NumberOfGCLogFiles=9 > -XX:OnOutOfMemoryError=/opt/solr6/bin/oom_solr.sh 8982 /index/solr6/logs > -Xloggc:/index/solr6/logs/solr_gc.log > -Xms28g > -Xmx28g > -Xss256k > -verbose:gc > > Neither system has any huge pages allocated in the OS, so I doubt that > the UseLargePages option is actually doing anything. I've left it there > in case I *do* enable huge pages, so they will automatically get used. > > Thanks, > Shawn > >
Re: Solr uses lots of shared memory!
On 8/23/2017 7:32 AM, Markus Jelsma wrote: > Why does it slowly increase over time? > Why does it appear to correlate to index size? > Is anyone else seeing this on their 6.6 cloud production or local machines? More detailed information included here. My 6.6 dev install is NOT having the problem, but a much older version IS. I grabbed this screenshot only moments ago from a production server which is exhibiting a large SHR value for the Solr process: https://www.dropbox.com/s/q79lo2gft9es06u/idxa1-top-big-shr.png?dl=0 This is Solr 4.7.2, with a 10 month uptime for the Solr process, running with these arguments: -DSTOP.KEY=REDACTED -DSTOP.PORT=8078 -Djetty.port=8981 -Dsolr.solr.home=/index/solr4 -Dcom.sun.management.jmxremote.authenticate=false -Dcom.sun.management.jmxremote.ssl=false -Dcom.sun.management.jmxremote.port=8686 -Dcom.sun.management.jmxremote -XX:+PrintReferenceGC -XX:+PrintAdaptiveSizePolicy -XX:+PrintGCDetails -XX:+PrintGCDateStamps -Xloggc:logs/gc.log-verbose:gc -XX:+AggressiveOpts -XX:+UseLargePages -XX:InitiatingHeapOccupancyPercent=75 -XX:MaxGCPauseMillis=250 -XX:G1HeapRegionSize=8m -XX:+ParallelRefProcEnabled -XX:+PerfDisableSharedMem -XX:+UseG1GC -Dlog4j.configuration=file:etc/log4j.properties -Xmx8192M -Xms4096M The OS is CentOS 6, with the following Java and kernel: java version "1.7.0_72" Java(TM) SE Runtime Environment (build 1.7.0_72-b14) Java HotSpot(TM) 64-Bit Server VM (build 24.72-b04, mixed mode) Linux idxa1 2.6.32-431.11.2.el6.centos.plus.x86_64 #1 SMP Tue Mar 25 21:36:54 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux I also just grabbed a screenshot from my dev server, running Ubuntu 14, Solr 6.6.0, a LOT more index data, and a more recent Java version. Solr has an uptime of about one month. This server was installed with the service installer script, so it uses bin/solr. It doesn't seem to have the same problem: https://www.dropbox.com/s/85h1weuopa643za/bigindy5-top-small-shr.png?dl=0 java version "1.8.0_144" Java(TM) SE Runtime Environment (build 1.8.0_144-b01) Java HotSpot(TM) 64-Bit Server VM (build 25.144-b01, mixed mode) Linux bigindy5 3.13.0-105-generic #152-Ubuntu SMP Fri Dec 2 15:37:11 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux The arguments for this one are very similar to the production server: -DSTOP.KEY=solrrocks -DSTOP.PORT=7982 -Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.authenticate=false -Dcom.sun.management.jmxremote.local.only=false -Dcom.sun.management.jmxremote.port=18982 -Dcom.sun.management.jmxremote.rmi.port=18982 -Dcom.sun.management.jmxremote.ssl=false -Djetty.home=/opt/solr6/server -Djetty.port=8982 -Dlog4j.configuration=file:/index/solr6/log4j.properties -Dsolr.install.dir=/opt/solr6 -Dsolr.log.dir=/index/solr6/logs -Dsolr.log.muteconsole -Dsolr.solr.home=/index/solr6/data -Duser.timezone=UTC -XX:+AggressiveOpts -XX:+ParallelRefProcEnabled -XX:+PrintGCApplicationStoppedTime -XX:+PrintGCDateStamps -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintHeapAtGC -XX:+PrintTenuringDistribution -XX:+UseG1GC -XX:+UseGCLogFileRotation -XX:+UseLargePages -XX:G1HeapRegionSize=8m -XX:GCLogFileSize=20M -XX:InitiatingHeapOccupancyPercent=75 -XX:MaxGCPauseMillis=250 -XX:NumberOfGCLogFiles=9 -XX:OnOutOfMemoryError=/opt/solr6/bin/oom_solr.sh 8982 /index/solr6/logs -Xloggc:/index/solr6/logs/solr_gc.log -Xms28g -Xmx28g -Xss256k -verbose:gc Neither system has any huge pages allocated in the OS, so I doubt that the UseLargePages option is actually doing anything. I've left it there in case I *do* enable huge pages, so they will automatically get used. Thanks, Shawn
RE: Solr uses lots of shared memory!
I do not think it is a problem of reporting after watching top after restart of some Solr instances, it dropped back to `normal`, around 350 MB, which i think it high to but anyway. Two hours later, the restarted nodes are slowly increasing shared memory consumption to about 1500 MB now. I don't understand why shared memory usage should/would increase slowly over time, it makes little sense to me and i cannot remember Solr doing this in the past ten years. But it seems to correlate to index size on disk, these main text search nodes have an index of around 16 GB and up 3 GB of shared memory after a few days. Logs nodes up to 800 MB index size and 320 MB of shared memory, the low latency nodes have four different cores that make up just over 100 MB index size, shared memory consumption is just 22 MB, which seems more reasonable for the case of shared memory. I can also force Solr to 'leak' shared memory just by sending queries to it. My freshly restarted local node used 68 MB shared memory at startup. Two minutes and 25.000 queries later it was already 2748 MB! At first there is a very sharp increase to 2000, then it takes almost two minutes more to increase to 2748. I can decrease the maximum shared memory usage to 1200 if i query (via edismax) only on fields of one language instead of 25 orso. I can decrease it as well further if i disable highlighting (HUH?) but still query on all fields. * We have tried patching Java's ByteBuffer [1] because it seemed to fit the problems, it does not fix it. * We have also removed all our custom plugins, so it has become a vanilla Solr 6.6 just with our stripped down schema and solrconfig, it neither fixes it. Why does it slowly increase over time? Why does it appear to correlate to index size? Is anyone else seeing this on their 6.6 cloud production or local machines? Thanks, Markus [1]: http://www.evanjones.ca/java-bytebuffer-leak.html -Original message- > From:Shawn Heisey > Sent: Tuesday 22nd August 2017 17:32 > To: solr-user@lucene.apache.org > Subject: Re: Solr uses lots of shared memory! > > On 8/22/2017 7:24 AM, Markus Jelsma wrote: > > I have never seen this before, one of our collections, all nodes eating > > tons of shared memory! > > > > Here's one of the nodes: > > 10497 solr 20 0 19.439g 4.505g 3.139g S 1.0 57.8 2511:46 java > > > > RSS is roughly equal to heap size + usual off-heap space + shared memory. > > Virtual is equal to RSS and index size on disk. For two other collections, > > the nodes use shared memory as expected, in the MB range. > > > > How can Solr, this collection, use so much shared memory? Why? > > I've seen this on my own servers at work, and when I add up a subset of > the memory numbers I can see from the system, it ends up being more > memory than I even have in the server. > > I suspect there is something odd going on in how Java reports memory > usage to the OS, or maybe a glitch in how Linux interprets Java's memory > usage. At some point in the past, numbers were reported correctly. I > do not know if the change came about because of a Solr upgrade, because > of a Java upgrade, or because of an OS kernel upgrade. All three were > upgraded between when I know the numbers looked right and when I noticed > they were wrong. > > https://www.dropbox.com/s/91uqlrnfghr2heo/solr-memory-sorted-top.png?dl=0 > > This screenshot shows that Solr is using 17GB of memory, 41.45GB of > memory is being used by the OS disk cache, and 10.23GB of memory is > free. Add those up, and it comes to 68.68GB ... but the machine only > has 64GB of memory, and that total doesn't include the memory usage of > the other processes seen in the screenshot. This impossible situation > means that something is being misreported somewhere. If I deduct that > 11GB of SHR from the RES value, then all the numbers work. > > The screenshot was almost 3 years ago, so I do not know what machine it > came from, and therefore I can't be sure what the actual heap size was. > I think it was about 6GB -- the difference between RES and SHR. I have > used a 6GB heap on some of my production servers in the past. The > server where I got this screenshot was not having any noticeable > performance or memory problems, so I think that I can trust that the > main numbers above the process list (which only come from the OS) are > correct. > > Thanks, > Shawn > >
Re: Solr uses lots of shared memory!
On 8/22/2017 7:24 AM, Markus Jelsma wrote: > I have never seen this before, one of our collections, all nodes eating tons > of shared memory! > > Here's one of the nodes: > 10497 solr 20 0 19.439g 4.505g 3.139g S 1.0 57.8 2511:46 java > > RSS is roughly equal to heap size + usual off-heap space + shared memory. > Virtual is equal to RSS and index size on disk. For two other collections, > the nodes use shared memory as expected, in the MB range. > > How can Solr, this collection, use so much shared memory? Why? I've seen this on my own servers at work, and when I add up a subset of the memory numbers I can see from the system, it ends up being more memory than I even have in the server. I suspect there is something odd going on in how Java reports memory usage to the OS, or maybe a glitch in how Linux interprets Java's memory usage. At some point in the past, numbers were reported correctly. I do not know if the change came about because of a Solr upgrade, because of a Java upgrade, or because of an OS kernel upgrade. All three were upgraded between when I know the numbers looked right and when I noticed they were wrong. https://www.dropbox.com/s/91uqlrnfghr2heo/solr-memory-sorted-top.png?dl=0 This screenshot shows that Solr is using 17GB of memory, 41.45GB of memory is being used by the OS disk cache, and 10.23GB of memory is free. Add those up, and it comes to 68.68GB ... but the machine only has 64GB of memory, and that total doesn't include the memory usage of the other processes seen in the screenshot. This impossible situation means that something is being misreported somewhere. If I deduct that 11GB of SHR from the RES value, then all the numbers work. The screenshot was almost 3 years ago, so I do not know what machine it came from, and therefore I can't be sure what the actual heap size was. I think it was about 6GB -- the difference between RES and SHR. I have used a 6GB heap on some of my production servers in the past. The server where I got this screenshot was not having any noticeable performance or memory problems, so I think that I can trust that the main numbers above the process list (which only come from the OS) are correct. Thanks, Shawn