Re: [vchkpw] vmysql: sql error[3]: MySQL server has gone away

2010-05-14 Thread Sérgio Rosa

Have a look at this link
http://dev.mysql.com/doc/refman/5.0/en/gone-away.html

you may find some help there.

Also look at this links
http://dev.mysql.com/doc/refman/5.0/en/error-log.html
and
http://dev.mysql.com/doc/refman/5.0/en/slow-query-log.html

try to enable slow-query-log. It might come up with an unknown  slow query .

when you have those errors if you log to the mysql server can you query 
anything to the db? or the server is dead?


Also take a look at this link
http://www.unix.com/web-programming-web-2-0-mashups/75618-mysql-tuning-tools-mysqltuner-pl-tuning-primer-sh.html

Come back with reports and news.

sergio rosa

On 13/05/2010 17:15, Joe Littleton wrote:


I tried your config for my.cnf but I am still having the same issue.

It appears that if mysql is restarted whatever authdaemon is looking 
at is gone. It's like there is a specific PID or something and once 
that PID is gone authdaemon believes mysql has stopped.


Hi!

I had a similar problem and it was being caused by excess tuning to 
mysql. Now I run my mysql with this my.cnf


The issue was not with vpopmail but with web applications. A few years 
ago I had a mysqld has gone away but it was because a small 
max_connection value.


max_connections = 75  what is your number of max 
conn?

key_buffer = 120M
max_allowed_packet = 24M
table_cache = 3072
sort_buffer_size = 2M
read_buffer_size = 2M
read_rnd_buffer_size = 4M
myisam_sort_buffer_size = 64M
thread_cache_size = 8
query_cache_size = 128M
query_cache_limit = 2M
join_buffer_size = 1M
thread_concurrency = 2
long_query_time = 5
log-slow-queries = /tmp/slow-queries.log
open_files_limit = 6500
max_heap_table_size = 64M
query_cache_min_res_unit = 2K
wait_timeout = 60 --- Watch this value it was giving 
my headache
connect_timeout = 10 --- Watch this value it was 
giving my headache
interactive_timeout = 120 --- Watch this value it was 
giving my headache
max_connect_errors = 1000 --- Watch this value it was 
giving my headache

expire_logs_days = 7
low_priority_updates = 1
max_join_size = 4294967295

This is a dedicated server with 2G, so the others values my not apply 
to your setup.


Hope that helps.
sergio rosa

--

**AWD***
*arquitectura web e design, lda
rua do moinho velho, 19 2ºdto
2655-242 ericeira

tlm +351 913 489 195
mail sergior...@awd.pt mailto:sergior...@awd.pt

url http://www.awd.pt

Disclaimer For Citizens National Bank

The information transmitted is intended only for the person or entity to which it 
is addressed and may contain confidential and/or privileged material. Any review, 
retransmission, dissemination or other use of, or taking of any action in reliance upon 
this information by persons or entities other than the intended recipient is prohibited. 
If you received this message in error, please contact the sender and delete the material 
from all computers.

This original email was sent to the internet for delivery at 13 May 2010 
12:15:43 -0400
   
 


--
*AWD
*arquitectura web e design, lda
rua do moinho velho, 19 2ºdto
2655-242 ericeira

tlm +351 913 489 195
mail sergior...@awd.pt mailto:sergior...@awd.pt
url http://www.awd.pt



!DSPAM:4bed23ae32711753472123!


Re: [vchkpw] Segmentation fault with 5.5

2010-05-14 Thread Martin Spuetz
Hello Matt,

thank you for your answer.

On 11.05.2010 23:52, Matt Brookings wrote:
 int main()
 {
 if (valias_select(info, rumtun.de) != NULL) {
 
 valias_select calls vget_assign, which in turn modifies the domain
 field for its purposes.  Because you're providing a pointer to a
 read-only portion of memory, that's one reason why you might be
 segfaulting.
 
 If you're going to write code that calls vpopmail API, be aware that
 much of vpopmail's API will clobber data you pass to it.

my first fault was to think that 5.5.0 is a stable release. I tried
5.4.30 and it's working again. My 2-lines c-code and the chkuser patch.

Greetings
Martin

!DSPAM:4bedb79832717422568130!