User Tools

Site Tools


cluster:100

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
cluster:100 [2011/03/27 17:38]
hmeij
cluster:100 [2011/03/28 10:23] (current)
hmeij
Line 4: Line 4:
 ==== Quotas ==== ==== Quotas ====
  
-03/​25/​2011 ​Quotas ​have been enabled on cluster greentail. ​ Because group quota numbers are weird, it is like counting duplicates or following links or looping, not sure, we have to rely on individual quotas. Just as well.+On 03/​25/​2011 ​quotas ​have been enabled on cluster greentail. ​ Because group quota numbers are weird, it is like counting duplicates or following links or looping, not sure, we have to rely on individual quotas. Just as well.
  
-Here is what was put in place, and will mature over time.  But first, it is important to understand:+Here is what was put in place, and will mature/​change ​over time.  But first, it is important to understand:
  
-  * A quota is not disk space allocated for you, as in reserved.  ​IT is simply a threshold (soft/low) that can be adjusted, or not. +  * A quota is not disk space allocated for you, as in reserved.  ​It is simply a threshold (soft/low) that can be adjusted, or not. 
-  * Reason for quotas: to avoid the /home to fill up by a runaway process, and constrain ​certain ​large users.+  * Reason for quotas: to avoid /home from filling ​up by a runaway process, and constrain large users from growing beyond arbitrary thresholds.
  
-The quotas fall into two tiers: the sliding tier and the custom tier.  ​In the sliding tier, when you reach your soft limit, notify HPC admins and your quota will slide to the next strata (<100G, <250G, <500G, <750G). Once you reach the last strata, and you still need more disk space, you enter the custom tier.  Here we set a quota after agreeing on size and availability of disk space, plus 100G.  Then you are on your own. +The quotas fall into two tiers: the "sliding" ​tier and the "custom" ​tier.  ​
  
-We urge all users to consider archiving data and files off the cluster ​disk array.+  * In the sliding tier, when you reach your soft limit, notify HPC admins and your quota will slide to the next strata (< 90/100G, < 225/250G, < 450/500G). Once you exceed the last strata, ​and you still need more disk space, you enter the custom tier.   
 +  * In the custom tier, we discuss setting a quota after agreeing on size and availability of disk space, plus 100G.  Then you are on your own. (Users that currently fall into this group have been grandfathered in and notified).
  
-How to find my quota?+We urge all users to consider archiving data and files off the cluster disk array. ​ Any quota can ofcourse **shrink**, please let us know if you move files off the array and free up disk space for others. 
 + 
 +How to find my quota? ​You can find your quota levels and current usage as follows. ​ Units are in kilobytes. So the blocks soft limit (quota) below is 90 GB, the hard limit (limit) is 100 GB.   
 + 
 +  * The hard block limit is the absolute maximum amount of disk space that a user can use. Once this limit is reached, no additional disk space can be used. Writes to disks will fail. 
 +  * The soft block limit defines the maximum amount of disk space that can be used. Unlike the hard limit, the soft limit can be exceeded for a certain amount of time. That time is known as the grace period. Warnings will appear in your terminal/​shell.  
 +  * The grace period is globally set for 7 days.  
 +  * We do not enforce inode limits. 
 + 
 +<​code>​ 
 + 
 +[hmeij@greentail ~]$ quota -v hmeij 
 + 
 +Disk quotas for user hmeij (uid 8216):  
 +     ​Filesystem ​ blocks ​  ​quota ​  ​limit ​  ​grace ​  ​files ​  ​quota ​  ​limit ​  ​grace 
 +/​dev/​mapper/​vghome-lvhome 
 +                1227284 ​ 90000000 100000000 ​           1020       ​0 ​      ​0 ​        
 + 
 +</​code>​
  
  
 ==== Notes ==== ==== Notes ====
  
-  ​* Implementation+**Implementation** 
   * add usrquota,​grpquota options to fstab, remount   * add usrquota,​grpquota options to fstab, remount
   * log everybody out or reboot (quota error on boot, not initialized yet)   * log everybody out or reboot (quota error on boot, not initialized yet)
Line 31: Line 51:
   * group quota numbers do not add up !?   * group quota numbers do not add up !?
  
-  ​* Array +**Array** 
-  * 48 TB minus 4 global spares + 
-  * after formatting 36 TB available +  * 48 one TB disk minus 4 global spares ​leave 44 disks 
-  * /home 10, /​snapshots ​10, /sanscratch 5 after that leaves 11 TB+  * after formatting 36 TB available ​(81%) 
 +  * /home 12 disks, /​snapshots ​12 disks, /​sanscratch ​24 disks (presents a TB /sanscratch volume) 
 +  * leaves ​roughly ​11 disks free in /​sanscratch,​ needed for /snapshot growth
   * snapshots holds base plus deltas hence needs to be somewhere between 1-2x base   * snapshots holds base plus deltas hence needs to be somewhere between 1-2x base
-  * so if we wanted to increase home, we could not use sanscratch (is anybody using it?)+  * so if we wanted to increase home, we delete ​sanscratch ​volume ​(is anybody using it?)
   * if we do not backup, then ... questions for later   * if we do not backup, then ... questions for later
  
-  ​* Links+**Links** 
   * [[http://​tldp.org/​HOWTO/​Quota-4.html|Quota mini HOWTO]]   * [[http://​tldp.org/​HOWTO/​Quota-4.html|Quota mini HOWTO]]
   * [[http://​www.linuxtopia.org/​online_books/​rhel5/​rhel5_administration/​rhel5_ch-disk-quotas.html|Admin:​ Quotas]]   * [[http://​www.linuxtopia.org/​online_books/​rhel5/​rhel5_administration/​rhel5_ch-disk-quotas.html|Admin:​ Quotas]]
cluster/100.1301261903.txt.gz · Last modified: 2011/03/27 17:38 by hmeij