![]() ![]() ![]() LEFT JOIN locked_field_bean_rel calls_locked_fields ON (calls. LEFT JOIN tags jt6_tag_link ON (jt6_tag_link. LEFT JOIN tag_bean_rel calls_tags ON (calls. LEFT JOIN users_cstm jt5_favorite_link_cstm ON jt5_favorite_link_cstm. LEFT JOIN users jt5_favorite_link ON (jt5_favorite_link. LEFT JOIN sugarfavorites calls_favorite ON (calls. LEFT JOIN users_cstm jt4_following_link_cstm ON jt4_following_link_cstm. LEFT JOIN users jt4_following_link ON (jt4_following_link. LEFT JOIN subscriptions calls_following ON (calls. LEFT JOIN contacts_cstm jt3_contacts_cstm ON jt3_contacts_cstm. LEFT JOIN contacts jt3_contacts ON (jt3_contacts. LEFT JOIN calls_contacts calls_contacts ON (calls. LEFT JOIN users_cstm jt2_created_by_link_cstm ON jt2_created_by_link_cstm. LEFT JOIN users jt2_created_by_link ON (calls. LEFT JOIN users_cstm jt1_modified_user_link_cstm ON jt1_modified_user_link_cstm. modified_user_id = jt1_modified_user_link. LEFT JOIN users jt1_modified_user_link ON (calls. LEFT JOIN users_cstm jt0_users_cstm ON jt0_users_cstm. LEFT JOIN calls_users calls_users ON (calls. INNER JOIN team_memberships team_memberships ON tst. assigned_user_id amp_pitchingassignments_accounts_name_owner , name amp_pitchingassignments_accounts_name , assigned_user_id amp_pitchingassignments_accountsamp_pitchingassignments_ida_owner , id amp_pitchingassignments_accountsamp_pitchingassignments_ida , last_name rel_assigned_user_name_last_name , first_name rel_assigned_user_name_first_name , first_name, '' ), ' ' ,IFNULL (jt3_contacts. LTRIM (RTRIM (CONCAT (IFNULL (jt3_contacts. last_name rel_created_by_name_last_name , first_name rel_created_by_name_first_name , last_name rel_modified_by_name_last_name , first_name rel_modified_by_name_first_name , Innodb_temp_data_file_path = ibtmp1:12M:autoextend:max:128Gīelow is one of the queries that gets stuck, I suspect it's a Sugar stock query for hydrating the Calls notifications a user gets in the UI periodically. We tried a couple of config changes (increasing buffers, table cache, tmp table size) but in the end the problem always comes back after a while. After restarting the database server things seem to be working fine for up to a few days.īelow is the database server config. It appears that the problem manifests itself only after some uptime of the MySQL instance. The data itself with indexes is around 16,9GB. The MySQL instance is housed on a separate virtual server with 8 CPU cores assigned and 32GB of RAM. The query execution gets stuck in the creating sort index phase, see the attached screen shot. The problem is caused by some of the sql queries taking an extremely long time which effectively causes many of them to get stacked. This in turn eats up the server resources causing everything to slow down. A client of ours is experiencing some weird slowness issue in Sugar 7.9 running on MySQL 5.7. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |