Re: Interesting PGA/Memory puzzle (4030 error at 32GB used) - ulimit settings?

From: Mladen Gogala <gogala.mladen_at_gmail.com>
Date: Fri, 18 May 2018 13:41:14 -0400
Message-ID: <9f44673f-e2b5-0db7-4bfc-3a3851cc8783_at_gmail.com>



It may be that you are running out of segments. Check max_map_count parameter and make sure it's 131072 or larger. The default is 65536, which can be too low for a busy database with many users. Even 262144 is not too outlandish.

Regards

On 05/18/2018 12:54 PM, Chris Taylor wrote:
> Also (if it matters) this is a JDBC connected program from ODI.
>
> Chris
>
> On Fri, May 18, 2018 at 11:53 AM, Chris Taylor
> <christopherdtaylor1994_at_gmail.com
> <mailto:christopherdtaylor1994_at_gmail.com>> wrote:
>
> Env: 12.1.0.2 on Linux 7.4 64-bit
>
> So, I've been all over Oracle support and Google-ing this morning
> trying to determine a root cause for this:
>
> Here's the error and its always 32GB (ulimit settings to follow). 
> Oracle doesn't have any support docs for KOLLRSZ.
>
> The reason I'm wondering about ulimits kicking in is because its
> always at 32GB.  I'm betting its a memory leak but I'd like to
> overcome this seeming 32GB barrier.
>
> [TOC00001]
> ORA-04030: out of process memory when trying to allocate 16328
> bytes (koh-kghu sessi,kollrsz)
>
> [TOC00001-END]
> [TOC00002]
> ========= Dump for incident 931836 (ORA 4030) ========
> [TOC00003]
> ----- Beginning of Customized Incident Dump(s) -----
> =======================================
> TOP 10 MEMORY USES FOR THIS PROCESS
> ---------------------------------------
>
> *** 2018-05-18 12:33:30.917
> *100%   32 GB, 2087947 chunks: "kollrsz                   "*
> *         koh-kghu sessi ds=0x7f8047e4f178  dsprt=0x7f804ae4cf60*
>  0% 9688 KB,  77 chunks: "kllcqas:kllsltba          "  SQL
>          QERHJ hash-joi ds=0x7f8048f2dc40  dsprt=0x7f804a5ab710
>  0% 4333 KB,  50 chunks: "permanent memory          "  SQL
>          kxs-heap-w ds=0x7f804a5ab710  dsprt=0x7f804ae4cf60
>  0% 2503 KB,  18 chunks: "QERHJ list array          "  SQL
>          QERHJ hash-joi ds=0x7f804a61dc30  dsprt=0x7f804a5ab710
>  0% 2432 KB,   7 chunks: "HT buckets             "  SQL
>          QERHJ hash-joi ds=0x7f804a61dc30  dsprt=0x7f804a5ab710
>  0% 2056 KB,   2 chunks: "kllcqgf:kllsltba          "
>          klcliti:kghds  ds=0x7f80491e9e10  dsprt=0x7f804f9a09e0
>  0% 1537 KB,  56 chunks: "QERHJ Bit vector          "  SQL
>          QERHJ hash-joi ds=0x7f8048f2dc40  dsprt=0x7f804a5ab710
>  0% 1381 KB,  66 chunks: "free memory              "  SQL
>          QERHJ hash-joi ds=0x7f804a61dc30  dsprt=0x7f804a5ab710
>  0% 1213 KB,  27 chunks: "free memory              "
>          top call heap  ds=0x7f804f9a1be0  dsprt=(nil)
>  0% 1046 KB,  73 chunks: "free memory              "
>          top uga heap ds=0x7f804f9a1e00  dsprt=(nil)
>
>
>
> I'm wondering if there's a parameter at the OS layer limiting the
> session to 32GB
>
> ulimit -a
> -----------------------------------------
> core file size         (blocks, -c) 0
> data seg size          (kbytes, -d) unlimited
> scheduling priority             (-e) 0
> file size          (blocks, -f) unlimited
> pending signals                 (-i) 16506448
> max locked memory       (kbytes, -l) 64
> max memory size         (kbytes, -m) unlimited
> open files                 (-n) 131072
> pipe size       (512 bytes, -p) 8
> POSIX message queues     (bytes, -q) 819200
> real-time priority              (-r) 0
> stack size         (kbytes, -s) 32768
> cpu time        (seconds, -t) unlimited
> max user processes              (-u) 131072
> virtual memory         (kbytes, -v) unlimited
> file locks                 (-x) unlimited
>
> PGA Settings are as follows on this Data Warehouse:
>
> NAME                        TYPE        VALUE
> ------------------------------------ ----------- ------------
> pga_aggregate_limit                 big integer 128G
> pga_aggregate_target                big integer 8G
>
>
>

-- 
Mladen Gogala
Database Consultant
Tel: (347) 321-1217


--
http://www.freelists.org/webpage/oracle-l
Received on Fri May 18 2018 - 19:41:14 CEST

Original text of this message