It's kind of scary how often serendipity plays it's part. See the thread titled:
"Build suggester in different directory (not /tmp)." Which basically says that the suggester is being built in /tmp which may be limited. And yes, that's where it gets built by default, although the thread I mentioned has some potential work-arounds and a suggestion to make this configurable that would have to be implemented by (you may see a Solr JIRA to that effect generated soon). Of course you could make your /tmp volume bigger, but that may not be easy. Erick On Wed, Dec 20, 2017 at 11:46 AM, Fiz Newyorker <fiznewy...@gmail.com> wrote: > Hi Shawn/Erick/Matt, > > I agree with you. When I execute the command df -h I am getting the > complete list of nfs mount info and Size and available space. I just > shared one liner out of it.. > > > One more thing I observed whenever I run suggest.build. > > *http://rn.com:8989/solr/LW_Data/ <http://rn.com:8989/solr/LW_Data/>* > suggest?suggest=true&suggest.build=true&suggest.dictionary=fuzzySuggester&wt=json&suggest.q=wills&&indent=on&suggest.cfq=memory > > > The following files are created in TMP folder of the Machine. This is > causing the Issue. > > 2017-12-20 18:20:08.280 INFO (qtp401424608-14) [ x:LW_Data] > o.a.s.h.c.SuggestComponent SuggestComponent prepare with : > suggest.build=true&indent=on&suggest.q=wills&suggest.count=10&suggest=true&suggest.dictionary=fuzzySuggester&wt=json > > 2017-12-20 18:20:08.280 INFO (qtp401424608-14) [ x:LW_Data] > o.a.s.s.s.SolrSuggester SolrSuggester.build(fuzzySuggester) > > 2017-12-20 18:25:25.451 ERROR (qtp401424608-14) [ x:LW_Data] > o.a.s.h.RequestHandlerBase java.io.IOException: No space left on device > > at sun.nio.ch.FileDispatcherImpl.write0(Native Method) > > at sun.nio.ch.FileDispatcherImpl.write(FileDispatcherImpl.java:60) > > since my tmp folder is just 2GB . > > > *My Question is Why does AutoSuggest build command creates File in tmp > folder ? Is this Usual behavior ? * > > > -rw-------. 1 razord raz 554379579 Dec 20 18:40 suggester_input_0.tmp > > -rw-------. 1 razord ran 107963828 Dec 20 18:40 suggester_sort_1.tmp > > -rw-------. 1 razord raz 106162743 Dec 20 18:40 suggester_sort_2.tmp > > -rw-------. 1 razord raz 105870752 Dec 20 18:40 suggester_sort_3.tmp > > -rw-------. 1 razord raz 105851613 Dec 20 18:41 suggester_sort_4.tmp > > -rw-------. 1 razord raw 87888769 Dec 20 18:41 suggester_sort_5.tmp > > > > Thanks > > FIZ. > > On Wed, Dec 20, 2017 at 9:12 AM, Matthew Roth <mgrot...@gmail.com> wrote: > >> Oh, this seems relevant to my recent post to the list. My problem is that >> the suggester's are first being built in /tmp and moved to /var. tmp has a >> total of 2g's free whereas /var has near 100G. >> >> Perhaps you are running into the same problem I am in this regard? How does >> your /tmp dir look when building? >> >> Matt >> >> >> On Wed, Dec 20, 2017 at 2:59 AM, Shawn Heisey <apa...@elyograg.org> wrote: >> >> > On 12/20/2017 12:21 AM, Fiz Newyorker wrote: >> > >> >> I tried df -h , during suggest.build command. >> >> >> >> Size. Used Avail Use% Mounted on >> >> >> >> 63G 17G 44G 28% /ngs/app >> >> >> > >> > That cannot be the entire output of that command. Here's what I get when >> > I do it: >> > >> > root@smeagol:~# df -h >> > Filesystem Size Used Avail Use% Mounted on >> > udev 12G 0 12G 0% /dev >> > tmpfs 2.4G 251M 2.2G 11% /run >> > /dev/sda5 220G 15G 194G 8% / >> > tmpfs 12G 412K 12G 1% /dev/shm >> > tmpfs 5.0M 0 5.0M 0% /run/lock >> > tmpfs 12G 0 12G 0% /sys/fs/cgroup >> > /dev/sda1 47G 248M 45G 1% /boot >> > tmpfs 2.4G 84K 2.4G 1% /run/user/1000 >> > tmpfs 2.4G 0 2.4G 0% /run/user/141 >> > tmpfs 2.4G 0 2.4G 0% /run/user/0 >> > >> > If the disk has enough free space, then there is probably something else >> > at work, like a filesystem quota for the user that is running Solr, or >> some >> > other kind of limitation that has been configured. >> > >> > Thanks, >> > Shawn >> > >>
