General

In most computing environments, disc space is not infinite. The diskquota system provides a mechanism to control usage of disc space, on an individual basis. Quotas may be set for each individual user, on any, or all filesystems. The quota system will warn users when they exceed their allotted limit, but allow some extra space for current work. Repeatedly remaining over quota at logout, will cause a fatal over quota condition eventually. The quota system is an optional part of LINUX that may be included when the system is configured. This document will describe from what view the Users will see the quota-system and the way the quota-system can be configured by the System Administator .

Users' view of diskquotas

General

To most users, diskquotas will either be of no concern, or a fact of life that cannot be avoided. The quota(1) command will provide information on any disc quotas that may have been imposed upon a user.

There are two individual possible quotas that may be imposed, usually if one is, both will be. A limit can be set on the amount of space a user can occupy, and there may be a limit on the number of files (inodes) he can own. Quota provides information on the quotas that have been set by the system administrators, in each of these areas, and current usage. The inode limit and block limit are impossed both on uid and on gid. So if there are group quotas you can be limited because the group can't allocate any more space even if you as user still can allocate the space.

There are four numbers for each limit, the current usage, soft limit (quota), hard limit, and time remaining bfore the softlimit is intepretted as a hard limit. The soft limit is the number of 1K blocks (or files) that the user is expected to remain below. Each time the user's usage goes past this limit, he will be warned. The hard limit cannot be exceeded. If a user's usage reaches this number, further requests for space (or attempts to create a file) will fail with an EDQUOT error, and the first time this occurs, a message will be written to the user's terminal. Only one message will be output, until space occupied is reduced below the limit, and reaches it again, in order to avoid continual noise from those programs that ignore write errors.

When a use exeeds his softlimit a timer is set that normaly expires wuthin 7 days (1 week). The user can remove files in this period to make sure he is under the softlimit again before the timer expires. When the timer expires the particular limit that has been exceeded will be treated as if the hard limit has been reached, and no more resources will be allocated to the user. The only way to reset this condition is to reduce usage below the softlimit.

Surviving when quota limit is reached

In most cases, the only way to recover from over quota conditions, is to abort whatever activity was in progress on the filesystem that has reached its limit, remove sufficient files to bring the limit back below quota, and retry the failed program.

However, if you are in the editor and a write fails because of an over quota situation, that is not a suitable course of action, as it is most likely that initially attempting to write the file will have truncated its previous contents, so should the editor be aborted without correctly writing the file not only will the recent changes be lost, but possibly much, or even all, of the data that previously existed.

There are several possible safe exits for a user caught in this situation. He may use the editor ! shell escape command to examine his file space, and remove surplus files. Alternatively, using csh, he may suspend the editor, remove some files, then resume it. A third possibility, is to write the file to some other filesystem (perhaps to a file on /tmp) where the user's quota has not been exceeded. Then after rectifying the quota situation, the file can be moved back to the filesystem it belongs on.

Administering the quota system

To set up and establish the diskquota system, there are several steps necessary to be performed by the system administrator. The following steps must be taken:

Kernel configuration

 Before you can use the quota-system you must compile a kernel with the quota-system enabled. This is done by answering yes to the Disk QUOTA support question when running . Then run and install the new kernel images as the one that is booted at boottime.

Electing filesystems

 When you have a kernel that supports quota you need to make a decision as to what filesystems need to have quotas applied. Usually, only filesystems that house users' home directories, or other user files, will need to be subjected to the quota system, though it may also prove useful to also include /usr if its writable by normal users.

To enable quotas on a certain filesystem one should edit the /etc/fstab file and add entries for usrquota and grpquota. Mine looks like : 

# device		directory	type	options				
/dev/hda1		/		ext2	defaults			
/dev/hda2		none		swap	sw
/dev/hda3 		/usr		ext2	defaults
/dev/hdb1 		/usr/users	ext2	defaults,usrquota,grpquota	
/dev/hdb2 		/usr/src	ext2	defaults,usrquota		
none			/proc		proc	defaults
The keyword "usrquota" in the options field of each fstab-entry turns on userquota for this device. The keyword "grpquota" in the options field turns on groupquota for the device. When you use the usrquota and grpquota options without the "=" option you quotafiles are located in the rootdir of each filesystem. A file called "quota.user" is used for userquota and a file called "quota.group" is used for groupquota.

You can also define your quotafile yourself. Something like "usrquota=/usr/adm/quotasrc.user" puts the quotafile in /usr/adm with the name quotasrc.user. Please be aware of the maximum lenght a line can have in your fstab, see mntent.h for a definition.

Enabling quotas

Periodically (certainly after each unclean reboot, and when quotas are first enabled for a filesystem), the records retained in the quota file should be checked for consistency with the actual number of blocks and files allocated to the user. The quotacheck(8) command can be used to accomplish this. It is not necessary to dismount the filesystem, or disable the quota system to run this command, though on active filesystems inaccurate results may occur. This does no real harm in most cases, another run of quotacheck when the filesystem is idle will certainly correct any inaccuracy.

To check the filesystem for the actual number of blocks used by a user run quotacheck -avug to install or update all the quotafiles.

The quotacheck program takes some time on large filesystems, but whith the new version it is quite acceptable on my machine. But when you are hacking the kernel, I recommend not to use it because it takes some time every time you have to reboot your machine. You also can also put it in you rc script and run it like you run fsck on your filesystems only when the fastreboot flag is not set. There is no support for parallel checking of filesystems.

Ok now one should have all the quotafiles one needs. Now you can add a line like: /usr/etc/quotaon -avug

to your /etc/rc. This is to turn the quotas on when you boot your machine. This is they way to go and not turn it on yourself any time when you boot your machine.

Checking a filesystem for quotas

The super-user may use the quota (1) command to examine the usage and quotas of any user, and the repquota (8) command may be used to check the usages and limits for all users on a filesystem. Just run quotacheck -avug and the quotafiles are updated automagicaly and also the tables that are currently used by the kernel. Watch for the "updating in core quotas" message of the quotacheck (8) program this says if it updates the in core quotas in the kernel.

I cannot state this enough the quotafile is build as (uid || gid * sizeof(struct dquot)) so when you have nobody as uid 65535 and nobody owns a file you get big quota files, lets say of about 2 Mb all filed with zero's for users that don't have quota. So please be aware of that and don't mail me about that. It isn't much of a problem because the file isn't that big really all 0 blocks are not allocated on the disk.

Specifing a quota for a user or group

To edit the quotas for various users we use the edquota (8) program. Now use edquota -u to edit user quotas and edquota -g to edit group quotas.

Edit only the numbers behind the soft and hard keywords. There are two lines for each filesystem that has quota turned on. Soft means the softlimit, if people or groups go over there softlimit they have some grace period to make sure they go under there softlimit.

The graceperiod can be changed with edquota -t and enter the number of days there. If they don't remove it within there graceperiod it is counted as a hardlimit. The hardlimit is the absolute maximum they can allocate, if they want more the files are truncated.

The one line that says blocks is the number of blocks one can allocate, the line that says inodes is the number of inodes (files/named pipes/devices/etc.) one can allocate.

Most of the time you have groups of users with the same quota. A quick way of editing the quota for all those users is change to the dir where there homedirs reside. Do a edquota for one of the users and change the quotas to the approriate values. This user becomes the so called prototype user or group for all the others. Then execute edquota -p prototypeusername * this should do the trick, all users now have the quota they need, now you could edit the ones that still need other values.

Checking quotas for a user or group

Run the quota program. The syntax for this program is : 
quota [-guqv]
quota [-qv] -u username ...
quota [-qv] -g groupname ...
Use -v to see the quotas on Use -q to only see filesystems on which you are over your softlimit or have reached your hardlimit.

The -g flags give you all quotas for the groups you are in (also additional groups).

Disabling quota for a user or group

When you want to disable quotas for a certain user use the quota editor edquota. Type edquota username | uid or edquota -g groupname | gid and set block softlimit and hardlimit, and the inode soft- and hardlimit to 0. This should disable the quota for that user an the user can allocate as many blocks and inodes as he/she wants.

Quotas on NFS-mounted disks

 To have quotas on NFS filesystems, you need to install quotas on the fileserver and not on the client. Clients can obtain quota information with the quota command which does a query to the rquotad (8) running on the fileserver from which you mount your NFS disks. So don't put any usrquota or grpquota flags in the flags for mounting NFS disks. Instead install quotas on your fileserver. And start the rpc.rquotad from your network rc-file.

Some implementation details

Diskquota usage and information is stored in a file on the filesystem that the quotas are to be applied to. Conventionally, this file is quota.user or quota.group in the root of the filesystem.

The data in the file comprises an array of structures, indexed by uid or gid, one structure for each user or group on the system (whether the user or group has a quota on this filesystem or not). If the uid or gid space is sparse, then the file may have holes in it, which would be lost by copying, so it is best to avoid this.

The system is informed of the existence of the quota file by the quotactl (2) system call. It then reads the quota entries for each user or group currently active, then for any files open owned by users who are not currently active. Each subsequent open of a file on the filesystem, will be accompanied by a pairing with its quota information. In most cases this information will be retained in core, either because the user who owns the file is running some process, because other files are open owned by the same user, or because some file (perhaps this one) was recently accessed. In memory, the quota information is kept hashed by uid or group and filesystem, and retained in an LRU chain so recently released data can be easily reclaimed. Information about those users whose last process has recently terminated is also retained in this way.

Each time a block is accessed or released, and each time an inode is allocated or freed, the quota system gets told about it, and in the case of allocations, gets the opportunity to object. Measurements have shown that the quota code uses a very small percentage of the system cpu time consumed in writing a new block to disk.