MattB's questions

Our production web server has gone down a few times over the course of the last half year. In the end, we've needed to contact the web host and have them restart as I'm unable to even SSH in. This appears to only affect the web server and not the MySQL database server which is separate. When it affects the web server, all hosted websites time out.

I'd like to examine web server optimization/corrections to get to the root of this issue. Any recommendations on how to proceed with that? I'm sure log files would play a role. I'm able to find my way around a Linux-based server and make needed changes, but would be interested in any tips I may not have thought of yet. It may be best for us to speak with an outside consultant as another option.

Thanks.

We recently moved from name-based hosting to individual IPs for an SSL implementation. Thus, each website now has its own IP. While www.domain1.com and www.domain2.com are being directed to the correct directories on the server, subdomains such as dev.domain1.com and dev.domain2.com are not. The subdomains also have their own IPs. Here's the structure of one of the subdomain's .conf files:

<VirtualHost 1.2.3.4:80>
    ServerAdmin [email protected]
    DocumentRoot /var/www/domain1-dev
    ServerName dev.domain1.com
    ServerAlias dev.domain2.com
    ErrorLog logs/dev.domain1.com-error.log
    CustomLog logs/dev.domain1.com-access.log common

    <Directory /var/www/domain1-dev>
      Options Indexes FollowSymLinks
      AllowOverride All
      Order allow,deny
      Allow from all
    </Directory>
</VirtualHost>

ServerName is currently commented out and UseCanonicalName is set to be off in the main httpd.conf file. DocumentRoot is set to domain1's path in this httpd.conf file, as is the directive. NameVirtualHost is commented out as we're now on ip-based hosting.

Any idea what could be wrong? :-( It's almost as if subdomain .conf files aren't being read, or else the main domain's settings are being used instead. In fact, any subdomain I type is showing the primary domain's website, even if its not defined under a .conf file.

Thanks much.

EDIT

Here's what I get when running /usr/sbin/httpd -S:

# /usr/sbin/httpd -S
httpd: apr_sockaddr_info_get() failed for the.servername.com
httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName
VirtualHost configuration:
1.2.3.5:80        www.domain2.com (/etc/httpd/conf.d/domain2.conf:10)
1.2.3.5:443       www.domain2.com (/etc/httpd/conf.d/ssl.conf:87)
1.2.3.3:80        www.domain1.com (/etc/httpd/conf.d/domain1.conf:1)
1.2.3.3:443       www.domain1.com (/etc/httpd/conf.d/ssl.conf:45)
1.2.3.6:80        dev.domain1.com (/etc/httpd/conf.d/domain1-dev.conf:1)
Syntax OK

EDIT

Here's the syntax for each domain:

<VirtualHost 1.2.3.3:80>
    ServerAdmin [email protected]
    DocumentRoot /var/www/html
    ServerName www.domain1.com
    ServerAlias domain1.com
    ErrorLog logs/www.domain1.com-error.log
    CustomLog logs/www..domain1.com-access.log common
</VirtualHost>

In the past we've used name-based virtual hosting in Apache. We recently converted websites to SSL and had to go the ip-based route. As a result, we currently have an instance that is set up as follows:

• www.domain.com using port 80
• dev.domain.com using port 8080

Both use the same IP. Is this scenario possible using Drupal multi-site functionality? While we find that dev.domain.com works and reads the correct "dev" database (using the dev settings), it reads theme files from the "www" site instead which is not what we want.

Is the culprit the dev's htaccess file? Apache is listening on 8080 and does use the proper DB settings, but just not the correct theme files. One other note: browsing dev.domain.com:8080 gives an error: "The page isn't redirecting properly". Should we just purchase a new IP address for the dev website, or would this still not help?

Any advice would be appreciated. Thanks.

I've downloaded an ADO.NET and ODBC connector from the MySQL website. I'd like to access a remote MySQL server from within BIDS with the goal of importing the remote data and populating a local SQL Server database.

While it looks like making the connection is relatively easy, does anyone know if it's possible to connect securely so that data is encrypted?

MySQL is currently set to a wait_timeout of 28800 on our server which I understand to be the default. I've been attempting to change this to 60 as recommended by a consultant, but the changes don't seem to take for long. Here's what I've tried while logged into Linux as root and MySQL as a database user:

1) Logging into MySQL and:

show variables like '%wait_timeout%';
set wait_timeout=60;

2) Editing the my.cnf file under /etc and adding the following line to both [mysqld] and [mysqld_safe]:

wait_timeout=60

When I try #1 and choose to show the wait_timeout variable right after, it shows correctly as 60. However, if I wait a minute or two, it reverts back to 28800 again. When I try #2, the option doesn't seem to take at all, even after restarting Apache and MySQL.

The my.cnf seems to be the only such version of the file that I can find.

I've tried the following, but no arguments show:

mysql --print-defaults

When restarting Apache, I've been using:

/usr/sbin/apachectl graceful

When restarting MySQL, I've been using:

/etc/init.d/mysqld restart

Any ideas? :-(

We currently use name-based virtual hosting for two websites on one IP. The DNS entries for both point to the same IP at a 3rd party web host.

We've purchased SSL certificates for both which means name-based virtual hosting is no longer an option for us. It looks like we'll need to do ip-based virtual hosting instead. Due to this, we also purchased an additional IP to use for the 2nd website.

I'd like to verify these are the correct steps for making the change with minimal downtime:

1) Add the new IP at the Linux level Copy /etc/sysconfig/network-scripts/ifcfg-eth0 to /etc/sysconfig/network-scripts/ifcfg-eth0:1 and edit accordingly for the new IP. Then do an "ifdown eth0" and "ifup eth0". Ensure the new IP shows using "/sbin/ifconfig".

2) Edit httpd.conf. Remove "NameVirtualHost" directive and revise "VirtualHost" to be IP-specific. Verify using "/usr/sbin/apachectl configtest" and restart Apache.

3) Change the DNS A record for the 2nd website to point to the new IP.

Some questions:

1) From what I understand, #1 wouldn't involve any downtime and the existing IP on eth0 would remain intact. Is that correct?

2) In #2 above, are we able to set things so that either the old IP or new IP still direct to the correct websites? It sounds like this would temporarily be a mixed environment. I'm thinking this could provide a seamless option while DNS propagates.

3) Would I need to revise any routing? The only thing changing is the addition of the new IP and having that correctly point to the 2nd website.

4) Anything I missed?

Thanks!

I'm using the following command to import data from a 160 MB file created on another server with mysqldump. The environment I'm importing to is running the latest WAMP Server with max_allowed_packet set to "16M".

C:\wamp\bin\mysql\mysql5.1.36\bin>mysql --max_allowed_packet=15M -u root -pXXX drupal < C:\prod.sql

ERROR 2006 (HY000) at line 644: MySQL server has gone away

This seems to happen consistently after approximately 39 seconds and 85.5 MB imported.

I read many suggestions on increasing max_allowed_packet in my.ini, so I did to 16M rather than the default of 1M. Yet the error I'm receiving persists no matter what I do. Any ideas?

Now solved in my.ini:

[wampmysqld]
port        = 3306
socket      = /tmp/mysql.sock
key_buffer = 384M
max_allowed_packet = 100M
table_cache = 4096
sort_buffer_size = 2M
net_buffer_length = 2M
read_buffer_size = 2M
read_rnd_buffer_size = 64M
myisam_sort_buffer_size = 64M
basedir=c:/wamp/bin/mysql/mysql5.1.36
log-error=c:/wamp/logs/mysql.log
datadir=c:/wamp/bin/mysql/mysql5.1.36/data
thread_cache_size = 8
query_cache_size = 32M

innodb_buffer_pool_size = 384M
innodb_additional_mem_pool_size = 20M
# Set .._log_file_size to 25 % of buffer pool size
innodb_log_file_size = 10M
innodb_log_buffer_size = 64M
innodb_flush_log_at_trx_commit = 1
innodb_lock_wait_timeout = 100

wait_timeout = 28800

[mysqldump]
quick
max_allowed_packet = 100M

[isamchk]
key_buffer = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M

[myisamchk]
key_buffer = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M