Category: Server Problems

[FreeBSD]MySQL server could not start

Posted on by

Today, I noticed that the MySQL server on my FreeBSD server is down for unknown reason, and today is Thanksgiving.

Initially, I tried to restart the MySQL server using the following command, but it didn’t help to restart the MySQL:

sudo /usr/local/etc/rc.d/mysql-server restart

and it gives me the following error message:

mysql not running? (check /var/db/mysql/icesquare.com.pid).

That’s interesting. So I check the pid file and I found that it is not available. After scratching my head for a while, I noticed one thing:

df

Filesystem     Size    Used   Avail Capacity  Mounted on
/dev/ad4s1d    7.7G    7.7G     -1M   100%    /var

Obviously, the /var directory is full. No wonder why the MySQL could not create any file, which explains why it stopped working.

Since the server is already up and running, it will be too late to increase the size of /var without reinstalling the entire system. The simplest way is to soft link the /var directory to somewhere else. Since a lot of system applications depends on /var, I don’t want to move the entire /var into a different places. So I decide to softlink /var/log into /usr/var/log.

The steps are easy:

sudo su
mkdir -p /usr/var/
mv /var/log /usr/var/
ln -s /usr/var/log /var/

To verify your work, it should look something like that:

ls /var

lrwxr-xr-x   1 root    wheel       13B Nov 22 11:48 log -> /usr/var/log/

df

Filesystem     Size    Used   Avail Capacity  Mounted on
/dev/ad4s1d    7.7G    1.1G    6.0G    15%    /var

Now, if you try to start the MySQL server again, it may or may not work. If you experience any difficulties, try to reboot the server. After the server is reboot, MySQL should be up and running again.

Lesson learned: Always create single partition. 🙂

–Derrick

Our sponsors:

FreeBSD.org Was Compromised. What Should You Do Next?

Posted on by

In case you are not aware this bad news. FreeBSD.org was compromised on September 17, 2012. Good news is that the compromise was caused by a human error (a stupid developer leaked a SSH private key somewhere), which has nothing to do with the code quality of the FreeBSD. 🙂 Bad news is that this may affect you if you use port. 🙁

The compromise is believed to have occurred due to the leak of an SSH key from a developer who legitimately had access to the machines in question, and was not due to any vulnerability or code exploit within FreeBSD.

We unfortunately cannot guarantee the integrity of any packages available for installation between 19th September 2012 and 11th November 2012, or of any ports compiled from trees obtained via any means other than through svn.freebsd.org or one of its mirrors. Although we have no evidence to suggest any tampering took place and believe such interference is unlikely, we have to recommend you consider reinstalling any machine from scratch, using trusted sources.

In short, the port tree was affected from September 19, 2012 to November 11, 2012. If you downloaded the ports in between this period, you are likely a victim of this incident. 🙁 Here is how to clean up the mess:

Stop using cvsup/csup to update your port. Use portsnap instead.

If you already used portsnap, do the following:

sudo portsnap fetch extract
sudo portsnap upgrade

If you have no idea what is portsnap, here is a quick tutorial:

First, install the portmaster and portupgrade. Port-Upgrade downloads the new ports to your server and Port-Master updates your applications based on the downloaded ports.

sudo pkg_add -r portupgrade portmaster

Download all ports. Notice that this is one-time work.

sudo portsnap fetch extract

If you decide to update your port tree, run the following:

sudo portsnap fetch update

Or if you like to run it in a cron job, run the following instead:

sudo nano /etc/crontab

#Run every three hours
0       */3     *       *       *       root    portsnap -I cron update && pkg_version -vIL= >/dev/null 2>&1

Now you have an updated port tree. Next you will need to update your applications based on your local port tree:

sudo portmaster -Dai

That’s it.

And remember, NEVER put your SSH private key in a public area.

–Derrick

Our sponsors:

[FreeBSD]The devel/pkg-config port moved to devel/pkgconf; Reason: pkg-config has been replace by pkgconf; pkgconf-0.8.5 conflicts with installed package(s): pkg-config-0.25_1

Posted on by

FreeBSD is designed for users who got lots of spare time. Every once a while, it will give you some troubles (or “snacks” in programmer language), which help you to burn some of your spare time. If you cannot handle those troubles, then FreeBSD is not for you.

Although FreeBSD is a rock solid operating system, it is not smart enough to tell whether the owners (i.e., me and you) got plenty of spare time or not.

Anyway, comparing to Windows Server, I think FreeBSD is considered an honor student.

Okay, let’s go back to the error message. When I tried to update FreeBSD, I saw the following error: 

===>>> All >> pkg-config-0.25_1 (1/1)

        ===>>> The devel/pkg-config port moved to devel/pkgconf
        ===>>> Reason: pkg-config has been replace by pkgconf


===>  pkgconf-0.8.5 conflicts with installed package(s):
      pkg-config-0.25_1

      They install files into the same place.
      You may want to stop build with Ctrl + C.

Again, this is an old-school error. The port-maintainer decided that the app A is too old, and he likes to replace it by app B. We all love the new stuffs, but we want the transition goes smooth. Apparently, it seems that the port-maintainer totally forgot about it.

So here is how to solve this problem:

sudo pkg_info | grep pkg

You will see something like the following:

pkg-config-0.25_1   A utility to retrieve information about installed libraries

Let’s delete it.

sudo pkg_delete -f pkg-config-0.25_1

Verify that the package has been deleted:

sudo pkg_info | grep pkg

Now try to install the package again:

cd /usr/ports/devel/pkgconf
sudo make install clean

That’s it!

–Derrick

Our sponsors:

[FreeBSD]How to run PECL Tokyo Tyrant in PHP 5.4

Posted on by

After upgrading to PHP 5.4.3 (which is the first version of PHP 5.4 available in the FreeBSD port), I found that the pecl-tokyo_tyrant could not be compiled. Here is the error messages:

cd /usr/ports/databases/pecl-tokyo_tyrant
sudo make

===>  pecl-tokyo_tyrant-0.6.0 cannot install: doesn't work with PHP version : 5 (Doesn't support PHP 5).
*** Error code 1

Stop in /usr/ports/databases/pecl-tokyo_tyrant.

/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:1827: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'php_tokyo_tyrant_class_methods'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:1867: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'php_tokyo_tyrant_table_class_methods'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:1911: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'php_tokyo_tyrant_query_class_methods'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:1936: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'php_tokyo_tyrant_iterator_class_methods'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c: In function 'php_tokyo_tyrant_query_object_new':
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2001: error: 'zend_class_entry' has no member named 'default_properties'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c: In function 'php_tokyo_tyrant_iterator_object_new':
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2052: error: 'zend_class_entry' has no member named 'default_properties'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c: In function 'php_tokyo_tyrant_object_new_ex':
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2090: error: 'zend_class_entry' has no member named 'default_properties'
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c: In function 'zm_startup_tokyo_tyrant':
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2166: error: 'php_tokyo_tyrant_class_methods' undeclared (first use in this function)
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2166: error: (Each undeclared identifier is reported only once
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2166: error: for each function it appears in.)
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2171: error: 'php_tokyo_tyrant_table_class_methods' undeclared (first use in this function)
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2176: error: 'php_tokyo_tyrant_query_class_methods' undeclared (first use in this function)
/usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0/tokyo_tyrant.c:2182: error: 'php_tokyo_tyrant_iterator_class_methods' undeclared (first use in this function)
*** Error code 1

Stop in /usr/ports/databases/pecl-tokyo_tyrant/work/tokyo_tyrant-0.6.0.
*** Error code 1

Stop in /usr/ports/databases/pecl-tokyo_tyrant.

Basically, it means that the pecl-tokyo_tyrant is not happy with the new version of PHP 5.4. I’ve tried several methods, including using pkg_add, compiling from source, copying a working tokyo_tyrant.so etc, unfortunately none of these methods works.

Finally, I found that the maintainer of this package has made a new version available(0.6.2) at github. Here is how I did it:

First, go to here to grab the source code. Make sure that you download the tag.gz version because it has the file attribute available.

Second, let’s remove the pecl-tokyo_tyrant from your system. You can either do it in the port or pkg_delete:

From port:

cd /usr/ports/databases/pecl-tokyo_tyrant
sudo make deinstall

Or using pkg_delete:

sudo pkg_delete pecl-tokyo_tyrant-0.6.0

Now, extract the files and compile it from source:

tar -zxvf mkoppanen-php-tokyo_tyrant-0.6.0-2-gb0d7e0c.tar.gz
cd mkoppanen-php-tokyo_tyrant-b0d7e0c
sudo phpize
sudo ./configure
sudo make
sudo make install

and don’t forget to include the .so file in your configuration:

sudo nano /usr/local/etc/php/extensions.ini

Add the following (if not available) to the end of the file:

extension=tokyo_tyrant.so

Finally, let’s restart the Apache:

sudo apachectl stop
sudo apachectl start

That’s it! Have fun with Tokyo Tyrant with PHP 5.4.

–Derrick

Our sponsors:

[PHP/FreeBSD]===>>> The databases/php5-sqlite port has been deleted: Removed from core php ===>>> Aborting update

Posted on by

When PHP 5.4 was available few months ago, I was hesitated to upgrade my FreeBSD box to upgrade to 5.4, because it comes with lots of big changes, including removing LOTS of old stuffs that were available in 5.3 or earlier versions.

Today PHP 5.4.3 becomes available in FreeBSD ports. I guess it is stable enough and I should give it a try, and I am glad that I tried it in my test machine first.

Warning: Before you upgrade to PHP 5.4, make sure that you are not using any of the following packages, because they are not compatible with the PHP 5.4. Unfortunately, I don’t have a solution yet. Basically, I could not get them compiled.

  • eAccelerator: /usr/ports/www/eaccelerator/
  • PECL-TokyoTyrant: /usr/ports/databases/pecl-tokyo_tyrant/ (Updated: I just came up a solution here)

When I performed the upgrade using the following command:

sudo portsnap fetch update
sudo portmaster -Da

I got the following error message:

===>>> The databases/php5-sqlite port has been deleted: Removed from core php
===>>> Aborting update

The reason why you see this problem because php5-sqlite is no longer available in PHP 5.4. In order to solve this problem, you will need to delete the php5-sqlite extension.

cd /usr/ports/lang/php5-extensions
sudo make config

Uncheck the following: sqlite3 support

Now, deinstall this port. Yes, let’s remove it and install it again. (I’ve tried the reinstall option and it didn’t work, so let’s stick with deinstall and install.

sudo make deinstall clean install clean

Now let’s find out which php-SQLite we should delete:

pkg_info | grep php5-sqlite

In my case, I deleted this one:

sudo pkg_delete php5-sqlite-5.3.13

Now upgrade the port again and FreeBSD should be happy about it. Enjoy PHP 5.4.

–Derrick

Our sponsors:

[Solved]Problem to Update OpenLDAP in FreeBSD

Posted on by

Recently, I found the following error message when updating my FreeBSD system regarding to the OpenLDAP:

===>  Installing for openldap-client-2.4.31
===>   Generating temporary packing list
===>  Checking if net/openldap24-client already installed
===> Creating users and/or groups.
Using existing group `ldap'.
Creating user `ldap' with uid `389'.
pw: user 'ldap' already exists
*** Error code 74

Stop in /usr/ports/net/openldap24-client.
*** Error code 1

Stop in /usr/ports/net/openldap24-client.

And I don’t use LDAP in particular, however it affects other applications that use LDAP, e.g., PHP

php -v
PHP Warning:  PHP Startup: Unable to load dynamic library '/usr/local/lib/php/20090626/ldap.so' - Shared object "libldap-2.4.so.8" not found, required by "ldap.so" in Unknown on line 0
PHP 5.3.10 with Suhosin-Patch (cli) (built: Feb 14 2012 16:59:15)
Copyright (c) 1997-2012 The PHP Group
Zend Engine v2.3.0, Copyright (c) 1998-2012 Zend Technologies
    with eAccelerator v0.9.6.1, Copyright (c) 2004-2010 eAccelerator, by eAccelerator

It is very simple to solve this problem. All you need is to remove the user ‘ldap’, and install the port again. First, let’s remove the user ‘ldap’ first:

sudo vipw

You will see this line:

ldap:*:389:389::0:0:OpenLDAP Server:/nonexistent:/usr/sbin/nologin

Remove the line by pressing ‘dd’, hit ‘ESC’ button, and then press ‘:wq’ to save the file and quit the editor.

After that, let’s clean the port and run the update again:

cd /usr/ports/net/openldap24-client
sudo make clean
sudo make install clean

That’s it. Your LDAP will be happy and your applications (such as PHP) will not complain.

Our sponsors:


–Derrick

[FreeBSD] Samba does not start after updating to 3.5.14

Posted on by

After I updated my Samba to 3.5.14 on my FreeBSD box today, it stopped working. I couldn’t connect to the Samba server from my Windows clients. Initially, I thought it was the memory cache issues, and I thought the problem would be solved by rebooting both machines. Unfortunately, it didn’t work. Therefore, I accessed to the server and check the cause of the problem.

First, I found that Samba was not evening running.

ps -ax | grep smb

So I started the Samba service manually:

sudo /usr/local/etc/rc.d/samba start

And it returned me the following message:

Cannot 'start' samba. Set winbindd_enable to YES in /etc/rc.conf or use 'onestart' instead of 'start'.

Oh that was easy. All I need to do was to modify my rc.conf:

#Samba service
samba_enable="YES"
winbindd_enable="YES"

And then I started the Samba again:

sudo /usr/local/etc/rc.d/samba start

Yay, Samba is up and running again!

–Derrick

Our sponsors:

How to Install a 2.5TB, 3TB drives on Linux

Posted on by

Recently I got a new hard drive (3TB) for my Linux box at work. I didn’t realize that it is problematic / nightmare to do such a simple work.

First, if you try to use fdisk to add the drive, it won’t work. Or I should say… it works, but it only give you a 2TB usable space. This is a limitation of fdisk, because it can handle up to 2^32 -1 sectors (or 2TB). That’s why fdisk is not a solution here.

It seems that the only solution is using parted.

sudo parted /dev/sdd

It will return something like this:

GNU Parted 2.3
Using /dev/sdd
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted)

First, we want to create a label:

(parted) mklabel gpt
Warning: The existing disk label on /dev/sdd will be destroyed and all data on this disk will be lost. Do you want to continue?
Yes/No? Yes

Then, we want to specify the unit:

(parted) unit TB

And tell parted to use all available space:

(parted) mkpart primary 0 -0

Next, we need to review the summary:

(parted) print
Model: ATA WDC WD30EZRX-00M (scsi)
Disk /dev/sdd: 3.00TB
Sector size (logical/physical): 512B/4096B
Partition Table: gpt

Number  Start   End     Size    File system  Name     Flags
 1      0.00TB  3.00TB  3.00TB  ext2         primary

As you can see, the size is 3TB. Don’t worry about the file system for now (ext2). If it looks good to you, now we need to leave parted.

(parted) quit
Information: You may need to update /etc/fstab.

Next, we need to format the partition:

#CentOS 6
sudo mkfs.ext4 /dev/sdd1

#CentOS 7
sudo mkfs.xfs /dev/sdd1

After few seconds, it will give you some more information about your drive:

mke2fs 1.41.14 (22-Dec-2010)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=1 blocks, Stripe width=1 blocks
183148544 inodes, 732566272 blocks
36628313 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
22357 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
        4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,
        102400000, 214990848, 512000000, 550731776, 644972544

Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done

This filesystem will be automatically checked every 35 mounts or
180 days, whichever comes first.  Use tune2fs -c or -i to override.

Now you can use your drive, such as mounting or putting it into /etc/fstab

sudo mount /dev/sdd1 /mnt/
df
/dev/sdd1             2.8T   42G  2.6T   2% /mnt

That’s it! Enjoy your new disk.

–Derrick

Our sponsors:

ZFS: Compression VS Deduplication (Dedup) in Simple English

Posted on by

Last Edited: Jan 17, 2021

Many people are confused between the ZFS compression and ZFS deduplication because they are so similar. Both of them are designed to reduce the size of the data being stored in the storage. Let me explain the difference between them in simple English.

1. This is how your data looks like originally (Assuming only one unique file):

2. This is how your data look like after being stored in a ZFS pool with compression enabled.

3. This is how your data look like after being stored in a ZFS pool with deduplication enabled.

4. Let say we are storing three identical files, i.e.,

5. ZFS: Compression Only

6. ZFS: Deduplication Only

7. ZFS: Compression + Deduplication

The biggest difference between deduplication and compression is the scope. File compression works at the file level. For example, if you have three identical files, ZFS will store the compressed files three times. Deduplication works at the block level. A block is simply the basic unit of the ZFS storage (e.g., 512 bytes, 4k etc). Imagine ZFS needs to store a big file. What it will do is to divide the file into multiple chunks. Each chunk will be stored in a block. What deduplication does is to remember the content of each block (checksum), and avoid storing the same content again. In other words, deduplication works at a narrower level (think of it as a molecule).

One of the reasons why the drug are usually tested on mice because some of mouse genes are 99% identical to human genes. Imagine we need to store the mouse genes into the database. All we need is to store the gene of mouse once. Later on if we need to store the human genes into the database, we can reference the mouse one rather than storing the same copy again.

Of course, enabling both compression and deduplication will save lots of free space. However, it comes with a very high price tag. If you like to enable deduplication, you need to make sure that you have at least 2GB of memory per 1TB of storage. For example, if your ZFS pool is 10TB, you need to have 20GB of memory installed in your system. Otherwise, you will experience a huge performance hit.

Hope this article helps you to understand the difference between compression and deduplication.

–Derrick

Our sponsors:

[Solved]Failed to enable the ‘dataready’ Accept Filter

Posted on by

After I updated the Apache to 2.2.22 for my FreeBSD box today, I got a problem:

#apachectl stop
#apachectl start

[warn] (2)No such file or directory: Failed to enable the 'dataready' Accept Filter

If you search for Failed to enable the ‘dataready’ Accept Filter on Google, you probably will find a lot of posts about this solution, like the following:

#kldload accf_http

Or include the following in the boot loader (/boot/loader.conf):

accf_http_load="YES"

However, even you already done these two things, this problem still exists. 

#kldload accf_http
kldload: can't load accf_http: File exists

Why? That’s because the problem is not coming from accf_http. Instead, the problem is the missing dataready filter, which is accf_data. To solve this problem, simply do the following. First, update the /boot/loader.conf and add the following into the file: 

accf_data_load="YES"

Of course, any changes to the boot loader requires a reboot. If you don’t want to reboot the machine, simple load the module manually and restart the Apache, i.e.,

#kldload accf_data
#apachectl stop
#apachectl start

That’s it! Apache will stop complaining the dataready filter and will work happily.

–Derrick

Our sponsors: