I'm going through a performance issue with a client and I am wondering if someone can help me interpret the processlist that I took a snapshot of today when the system was running slow for about 45 seconds for everyone. Here is what I got:
| Id | User | Host | db | Command | Time | State | Info |
| 367154 | sugarcrm | localhost | sugarcrm | Query | 98 | Locked | UPDATE users_cstm SET user_laston_c='XXX' ,user_module_c='Home' ,user_ip_c=' |
| 367158 | sugarcrm | localhost | sugarcrm | Query | 98 | Locked | UPDATE users_cstm SET user_laston_c='XXX' ,user_module_c='Prospects' ,user_ip_c=' |
| 367165 | sugarcrm | localhost | sugarcrm | Query | 28 | Locked | SELECT users.* , users_cstm.* FROM users LEFT JOIN users_cstm ON users.id = users_cstm.id_c WHERE |
| 367186 | root | localhost | sugarcrm | Query | 104 | Writing to net | SELECT /*!40001 SQL_NO_CACHE */ * FROM `emails` |
| 367187 | sugarcrm | localhost | sugarcrm | Query | 98 | Locked | SELECT users.* , users_cstm.* FROM users LEFT JOIN users_cstm ON users.id = users_cstm.id_c WHERE |
| 367188 | sugarcrm | localhost | sugarcrm | Query | 38 | Locked | SELECT users.* , users_cstm.* FROM users LEFT JOIN users_cstm ON users.id = users_cstm.id_c WHERE |

The one query that has a state of "Writing to net" confuses me slightly. First of all it is running under the root login and not the sugarcrm login, so I don't know what kicked it off, but the query itself does take awhile to run since there are over 77,000 emails in that table and they can be quite lengthy. The rest of the queries are locked, but I am not sure what has them locked, wouldn't at least one of them show as running and the other locked, because I don't think that the emails query should lock the tables, only the two update query would seem to lock the table.

The server has a single core opteron processor, and I'd like to see a dual-core, but it is what it is.

The tables are all set to MyISAM, and I watched the performance screencast on the Wiki and it said that they have better luck with InnoDB since it uses row locking instead of table locking, so I plan on making that change, which made resolve this speed issue, but I'm just wondering if anyone can help me understand what kicked off the emails query and why all of the other queries are locked instead of all but one.