內核不中斷前提下，Gaussdb（DWS）內存報錯排查方法

----debug_query_id=76279718689098154, memory allocation failed due to reaching the database memory limitation. Current thread is consuming about 10 MB, allocating 240064 bytes.
----debug_query_id=76279718689098154, Memory information of whole process in MB:max_dynamic_memory: 18770, dynamic_used_memory: 18770, dynamic_peak_memory: 18770, dynamic_used_shrctx: 1804, dynamic_peak_shrctx: 1826, max_sctpcomm_memory: 4000, sctpcomm_used_memory: 1786, sctpcomm_peak_memory: 1786, comm_global_memctx: 0, gpu_max_dynamic_memory: 0, gpu_dynamic_used_memory: 0, gpu_dynamic_peak_memory: 0, large_storage_memory: 0, process_used_memory: 22105, cstore_used_memory: 1022, shared_used_memory: 2605, other_used_memory: 0, os_totalmem: 257906, os_freemem: 16762.

----debug_query_id=72339069014641088, sessId: 1670914731.140604465997568.coordinator1, sessType: postgres, contextName: ExprContext, level: 5, parent: FunctionScan_140604465997568, totalSize: 950010640, freeSize: 0, usedSize: 950010640
----debug_query_id=72339069014641053, pid=140604465997568, application_name=gsql, query=select * from pv_total_memory_detail, state=retrying, query_start=2022-12-13 14:59:22.059805+08, enqueue=no waiting queue, connection_info={"driver_name":"gsql","driver_version":"(GaussDB 8.2.0 build bc4cec20) compiled at 2022-12-13 14:45:14 commit 3629 last mr 5138 debug","driver_path":"/data3/x00574567/self/gaussdb/mppdb_temp_install/bin/gsql","os_user":"x00574567"}
----debug_query_id=72339069014641088, sessId: 1670914731.140604738627328.coordinator1, sessType: postgres, contextName: ExprContext, level: 5, parent: FunctionScan_140604738627328, totalSize: 900010080, freeSize: 0, usedSize: 900010080
----debug_query_id=72339069014641057, pid=140604738627328, application_name=gsql, query=select * from pv_total_memory_detail, state=retrying, query_start=2022-12-13 14:59:22.098775+08, enqueue=no waiting queue, connection_info={"driver_name":"gsql","driver_version":"(GaussDB 8.2.0 build bc4cec20) compiled at 2022-12-13 14:45:14 commit 3629 last mr 5138 debug","driver_path":"/data3/x00574567/self/gaussdb/mppdb_temp_install/bin/gsql","os_user":"x00574567"}
----debug_query_id=72339069014641088, sessId: 1670914731.140603779163904.coordinator1, sessType: postgres, contextName: ExprContext, level: 5, parent: FunctionScan_140603779163904, totalSize: 890009968, freeSize: 0, usedSize: 890009968
----debug_query_id=72339069014641058, pid=140603779163904, application_name=gsql, query=select * from pv_total_memory_detail, state=retrying, query_start=2022-12-13 14:59:22.117463+08, enqueue=no waiting queue, connection_info={"driver_name":"gsql","driver_version":"(GaussDB 8.2.0 build bc4cec20) compiled at 2022-12-13 14:45:14 commit 3629 last mr 5138 debug","driver_path":"/data3/x00574567/self/gaussdb/mppdb_temp_install/bin/gsql","os_user":"x00574567"}
----allBackendSize=34, idleSize=7, runningSize=7, retryingSize=20

----debug_query_id=76279718689098154, Total estimated Memory is 15196 MB, total current cost Memory is 16454 MB, the difference is 1258 MB.The count of complicated queries is 17 and the count of uncontrolled queries is 1.

----debug_query_id=76279718689098154, The abnormal query thread id 140664667547392.It current used memory is 13618 MB and estimated memory is 1102 MB.It also is the query which costs the maximum memory.

----debug_query_id=76279718689098154, It is not the current session and beentry info : datid<16389>, app_name<cn_5001>, query_id<76279718688746485>, tid<140664667547392>, lwtid<173496>, parent_tid<0>, thread_level<0>, query_string<explainperformance with ws as (select d_year AS ws_sold_year, ws_item_sk, ws_bill_customer_sk ws_customer_sk, sum(ws_quantity) ws_qty, sum(ws_wholesale_cost) ws_wc, sum(ws_sales_price) ws_sp from web_sales left join web_returns on wr_order_number=ws_order_number and ws_item_sk=wr_item_sk join date_dim on ws_sold_date_sk = d_date_sk where wr_order_number is null group by d_year, ws_item_sk, ws_bill_customer_sk ), cs as (select d_year AS cs_sold_year, cs_item_sk, cs_bill_customer_sk cs_customer_sk, sum(cs_quantity) cs_qty, sum(cs_wholesale_cost) cs_wc, sum(cs_sales_price) cs_sp from catalog_sales left join catalog_returns on cr_order_number=cs_order_number and cs_item_sk=cr_item_sk join date_dim on cs_sold_date_sk =d_date_sk where cr_order_number is null group by d_year, cs_item_sk, cs_bill_customer_sk ), ss as (select d_year AS ss_sold_year, ss_item_sk, ss_customer_sk, sum(ss_quantity) ss_qty, sum(ss_wholesale_cost) ss_wc, sum(ss_sales_price) ss_spfrom store_sales left join store_returns on sr_ticket_numbe>.

----debug_query_id=76279718689098154, WARNING: the common memory context 'HashContext' is using 1059 MB size larger than 989 MB.----debug_query_id=76279718689098154, WARNING: the common memory context 'VecHashJoin_76279718688746485_6' is using 12359 MB size larger than 10 MB.

----debug_query_id=%lu, FATAL: memory allocation failed due to reaching the OS memory limitation. Current thread is consuming about %d MB, allocating %ld bytes.
----debug_query_id=%lu, Please check the sysctl configuration and GUC variable max_process_memory.
----debug_query_id=%lu, Memory information of whole process in MB:"
                            "max_dynamic_memory: %d, dynamic_used_memory: %d,
dynamic_peak_memory: %d, dynamic_used_shrctx: %d,
dynamic_peak_shrctx: %d, max_sctpcomm_memory: %d,
sctpcomm_used_memory: %d, sctpcomm_peak_memory: %d,
comm_global_memctx: %d, gpu_max_dynamic_memory: %d,
gpu_dynamic_used_memory: %d,
gpu_dynamic_peak_memory: %d, large_storage_memory: %d,
process_used_memory: %d, cstore_used_memory: %d,
shared_used_memory: %d, other_used_memory: %d,
os_totalmem: %d, os_freemem: %d

with a as (select *from pgxc_total_memory_detail where memorytype='dynamic_used_memory'), b as(select * from pgxc_total_memory_detail wherememorytype='dynamic_peak_memory'), c as (select * from pgxc_total_memory_detailwhere memorytype='max_dynamic_memory'), d as (select * frompgxc_total_memory_detail where memorytype='process_used_memory'), e as (select* from pgxc_total_memory_detail where memorytype='other_used_memory'), f as(select * from pgxc_total_memory_detail where memorytype='max_process_memory')select a.nodename,a.memorymbytes as dynamic_used_memory,b.memorymbytes asdynamic_peak_memory,c.memorymbytes as max_dynamic_memory,d.memorymbytes asprocess_used_memory,e.memorymbytes as other_used_memory,f.memorymbytes asmax_process_memory from a,b,c,d,e,f where a.nodename=b.nodename andb.nodename=c.nodename and c.nodename=d.nodename and d.nodename=e.nodename ande.nodename=f.nodename order by a.nodename;

SELECT * FROM pv_session_memory_detail ORDER BY totalsize desc LIMIT 100;

select sessid, contextname, level,parent, pg_size_pretty(totalsize) as total ,pg_size_pretty(freesize) as freesize, pg_size_pretty(usedsize) as usedsize, datname,query_id, query from pv_session_memory_detail a , pg_stat_activity b where split_part(a.sessid,'.',2) = b.pid order by totalsize desc limit 100;

EXECUTE DIRECT ON(cn_5001) 'SELECT pg_terminate_backend(139780156290816)';

select b.state, sum(totalsize) as totalsize, sum(freesize) as freesize, sum(usedsize) as usedsize from pv_session_memory_detail a , pg_stat_activity b where split_part(a.sessid,'.',2) = b.pid group by b.state order by totalsize desc limit 100;

CLEAN CONNECTION TO ALL FORCE FOR DATABASE xxxx;

select 'execute direct on ('||coorname||') ''select pg_terminate_backend('||pid||')'';' from pgxc_stat_activity where usename not in ('Ruby', 'omm') and state='idle';

select b.state as state, a.sessid as sessid, b.query_id as query_id, substr(b.query,1,100) as query, sum(totalsize) as totalsize, sum(freesize) as freesize, sum(usedsize) as usedsize from pv_session_memory_detail a , pg_stat_activity b where split_part(a.sessid,'.',2) = b.pid and usename not in ('Ruby', 'omm') group by state,sessid,query_id,query order by totalsize desc limit 100;

select * from pg_shared_memory_detail order by totalsize desc limit 10;