blunders's questions

"Ending company IT Admin relationship" has a good checklist for taking over an existing IT system, but I'm wondering as it relates to Linux:

• What is the most effective way to assess the scope of existing custom configurations, installs, scripts, etc done?
• Is there any software that will check if the kernel, system files, etc mirror the default files for the version installed?

At this point I don't know what distro of Linux the server (though using Netcraft I do know the server appears to be Linux) -- so it's possible without knowing that information that this would be a hard question to answer.

Researching legal and regulatory compliance for creating systems are tamper resistant, and wondering if there is a term for a system that is "Create, Read" enabled, but "update, delete" disabled.

Additionally, is two-factor authentication in general less secure, and if so, why?

An example of this would be something like the Google Search Appliance.

Areas covered might including closure and software security, remote access and updates, etc.

Looking more for a solution that covers 80% of the risks and requirements, than being at the Google Search Appliance level.

What is the difference between setting the max_allowed_packet via the MySQL commandline using:

set global max_allowed_packet=100000000;

And setting it by editing the MySQL my.ini (Win) or my.cnf (Linux) config and putting the following code under [mysqld]:

max_allowed_packet=100M 

Attempt to see the "net_buffer_length" config before resetting it:

mysql> show variables like "net_buffer_length";
+-------------------+-------+
| Variable_name     | Value |
+-------------------+-------+
| net_buffer_length | 16384 |
+-------------------+-------+
1 row in set (0.00 sec)

Attempt to reset "net_buffer_length" config:

mysql> set global net_buffer_length=1000000;
Query OK, 0 rows affected (0.00 sec)

Attempt to confirm the "net_buffer_length" config has been reset:

mysql> show variables like "net_buffer_length";
+-------------------+-------+
| Variable_name     | Value |
+-------------------+-------+
| net_buffer_length | 16384 |
+-------------------+-------+
1 row in set (0.00 sec)

What's wrong with the commands I'm using that result in the config not updating?

• MySQL Server Version: 5.1.53-community
• DATABASE_ENGINE: INNOdb

Questions, feedback, requests -- just comment, thanks!

Keep running into into the "max_allowed_packet" error msg, which to data I've just guessed at what the number should be....

I'd like to stop guessing, and know how to see what the number should be -- and it appears that info might be in the database table info in the "Data_length" column.

Is that correct, and if so, how do I calculate the value to enter in the MySQL config to avoid the error before I get it.

(If it matters, I keep getting these errors on DB restores.)

Questions, feedback, requests -- just comment, thanks!!

Additional info: The database is INNOdb, not MyISAM.

UPDATE (1):

In an effort to find a way to benchmark the "max_packet" I did the following, ran the restore, got the error with the line number causing the error, dumped that line into a text file to measure the file size.

ERROR:

ERROR 1153 (09S01) at line 2742: Got a packet bigger than "max_allowed_packet" bytes

SED code:

sed 'INSERTLINENUMBERq;d' INSERTDATABASEDUMPFILENAME.sql > INSERTLINENUMBER.txt

Problem is... that line is only 8mb, "max_allowed_packet" is set to 100mb.

As a follow-up to this question, I'm attempting to edit the "my.ini" config file for the MySQL service. I've stopped the service, opened the file, made the edit, and attempt to save it -- at which point I get an error saying "Access Denied". On Windows 7.

Questions, feedback, suggestions -- just comment, thanks!!

Getting the following error:

ERROR 2006 (HY000) at line 530: MySQL server has gone away

Which I get after running the following from the MySQL (5.1.53-community) command line:

CREATE DATABASE restoredb;

then this from the Windows 7 system command line:

mysql -u root -p restoredb < "C:\restoredb.sql"

I've tried another database restore using another file and the MySQL restore command works fine.

Questions, feedback, requests -- just comment, thanks!!

Trying to kill a process that running a MySQL database backup, and has the tables locked. What will happen if I kill the process?

What's the easy way to do this?

Problem is new to LAMP, and want to test the system config in any automated checklist way.

Any suggestions?

Examples:

• Was is the operating environment?
• How large of a dataset?
• Are offline backups acceptable?
• Is point-in-time restoration a requirement?
• What storage engine?

CONTEXT: Want to ask how to do MySQL backups on CentOS, but want to make sure I understand what I should be asking first. If you have any questions, just ask -- thanks!!

UPDATE: Just a note to say thanks for ALL the great input so far, really glad I asked. Again, thanks!!

Here's what I've done so far

[username@VMname /]$ sudo /etc/init.d/httpd restart
[username@VMname /]$ sudo /etc/init.d/httpd restart
Stopping httpd:                                            [  OK  ]
Starting httpd:                                            [  OK  ]
[username@VMname /]$ /etc/init.d/httpd status
httpd (pid  23043) is running...

Then tried running...

tail -f /var/log/httpd/error.log

That returned file not found, so I looked and this is what I see:

[username@VMname /]$ cd /var/log/httpd
access_log  access_log.1  error_log  error_log.1  ssl_access_log  ssl_error_log  ssl_error_log.1  ssl_request_log

Any help is much appreciated. Thanks in advance!!

Trying to run the command "apache2ctl restart" and just got this message "apache2ctl:: command not found" -- thing is, Apache IS installed, running, and I'm actively doing development on the server right now... any ideas?

CONTEXT: sudo access on CentOS-5.5 on VM

UPDATE: Good news, restarted Apache, bad news 500 error, still it's progress... :-)

[username@VMname /]$ sudo /etc/init.d/httpd restart
[username@VMname /]$ sudo /etc/init.d/httpd restart
Stopping httpd:                                            [  OK  ]
Starting httpd:                                            [  OK  ]

Trying to restart Apache, and can't tell if it's restarting -- are there logs for the restart, or if I input the commandline to restart should I expect a response (error msg, confirmation, status, etc.)

Never done it before and not sure what's going on.

This is exactly what I'm typing, no $/# before it...

COMMANDS INPUT:

sudo apache2ctl restart

-- or --

apache2ctl restart

CONTEXT: Commands executed from the root directory, meaning... "cd /"

RESULT: Nothing appears to happen.

Been dealing with growing demands for system resources on a website, and got me thinking:

What's the fastest way to server a single static webpage from a single server in memory?

Feel free to ask for clarifications, basically trying to reduce the application stack as much as possible, use as little memory as possible, but keep the system in memory as much as possible. Guessing the static webpage would not over https and would be 20k, no images, and inline CSS/JS; meaning it's one file. More points for lower cost/resource systems.

Looking for statistics on IPtable best pratices - appears that a server like www.cloudflare.com would parse iptables and look for gaps and trends in iptable configs - but don't think they'd publish their findings. FYI: I just heard about cloudflare, haven't used it, and have nothing to do with them. If you're aware of other services like cloudflare, please post links.

Have a CentOS-5.x box that appears to be going down as a result of PINGs. If the server gets PINGs it stops replying to ssh/http connections. Appears iptable is one approach, but I want to make sure I'm doing everything I should do.

QUESTION: What Command(s) List a User's System Permissions in Linux?

CURRENT_OS: CentOS-5.X

MORE_INFO/QUESTIONS:

Looking for commands that will NOT be seen as aggressive or change the system, meaning that I attempt to change the system OR access file/dir that have standard permissions assigned to XYZ role to "see" if it's possible to do; root, su/sudo, etc.

Which distro have the same command syntax? For example: (RHEL / Red Hat / CentOS / Fedora), (Debian / Ubuntu Linux), BSD, etc. -- NOTE: This list is NOT in any order, it's just a example, and I'm not looking for comments on which distro is better, just the commands requested.

Thank you!!

_______________

RECENT_UPDATE (1): I want to know if the user I'm currently signed in as has rights as root, su/sudo, etc. -- I'm not a Linux person, if I was on WIN-XP the answer would be to go to "START >> CONTROL PANEL >> USER ACCOUNTS" and I would see the system rights that user has; Admin, limited, etc. There other ways, but that's just an example. If there is a better way to ask the question, let me know. What I'm NOT looking for is what permissions I have on a file/folder/etc, I want to type a command-line and get info on the current users system rights, if that makes sense. Thanks!!

RECENT_UPDATE (2): Rewrote the question, posted it to stack-x-superuser, getting better responses: https://superuser.com/questions/197878/how-to-i-document-user-rights-on-an-existing-linux-system