Changes from Version 1 of YumMultipleMachineCaching

Show
Ignore:
Author:
james (IP: 65.172.155.230)
Timestamp:
12/17/08 18:21:53 (9 years ago)
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • YumMultipleMachineCaching

    v0 v1  
     1= Caching remote data for multiple computers = 
     2 
     3One of the things most people want is a way to not have to download the data yum/preupgrade/etc. needs multiple times over their connection to the internet. Speed is the major motivator, but also some people pay per. byte and so cost is a factor. This page lists the "best practice" available options, and their pros/cons: 
     4 
     5 * Create a mirror, using rsync/reposync, and point your computers at that. 
     6  * Pros 
     7   * Post setup it "just works" and everything is immediate. 
     8  * Cons 
     9   * Requires setting up a local webserver to serve the data. 
     10   * Requires syncing the data. 
     11   * Requires altering each computer to point to your local mirror. 
     12   * Requires downloading everything. 
     13   * preupgrade/etc. doesn't easily share data with "normal" yum, even if they need the same data. 
     14 
     15 * Create a mirror, using rsync/reposync, and register with MirrorManager. 
     16  * Pros 
     17   * Zero configuration, on the client side. 
     18   * Post setup it mostly just works and everything should be immediate. 
     19  * Cons 
     20   * Requires setting up a local webserver to serve the data. 
     21   * Requires syncing the data. 
     22   * If the server is down or the data becomes "out of date" then metalinks/MM will route you to an external mirror. 
     23   * Requires downloading everything. 
     24   * preupgrade/etc. doesn't easily share data with "normal" yum, even if they need the same data. 
     25 
     26 * Install "IntelligentMirror", and register with MirrorManager. 
     27  * Pros 
     28   * Zero configuration, on the client side. 
     29   * Post setup it mostly just works and everything should be immediate. 
     30   * Only downloads what is required by the users. 
     31   * Fully automated server side, once setup/working. 
     32  * Cons 
     33   * Requires setting up a local Squid + Apache-httpd + IntelligentMirror to serve the data. 
     34   * If the server is down then metalinks/MM will route you to an external mirror. 
     35   * Only intelligently caches packages, not metadata? 
     36 
     37 * Mount /var/cache/yum over NFS and set keepcache=1 in yum. 
     38  * Pros 
     39   * No server side. 
     40   * Only downloads what is required by the users. 
     41  * Cons 
     42   * sqlite (and hence yum) doesn't like being on NFS. 
     43   * running yum in parrallel (over any of the computers) is a bad idea. 
     44   * NFS. 
     45   * Requires setting up NFS mounts on each computer. 
     46   * Requires altering each computer to point to your local mirror. 
     47   * Cache will grow "without bound". 
     48   * preupgrade/etc. doesn't easily share data with "normal" yum, even if they need the same data. 
     49 
     50 * rsync /var/cache/yum and set keepcache=1 in yum. 
     51  * Pros 
     52   * No server side. 
     53   * Only downloads what is required by the users. 
     54  * Cons 
     55   * Requires running the sync before/after each use of yum that downloads anything. 
     56   * Requires altering each computer to point to your local mirror. 
     57   * Cache will grow "without bound". 
     58   * preupgrade/etc. doesn't easily share data with "normal" yum, even if they need the same data. 
     59 
     60 * (beta) avahi-packages-server.py and yum avahi support (see: http://james.fedorapeople.org/yum) 
     61  * Pros 
     62   * Zero configuration client. 
     63   * Zero configuration server. 
     64   * Only downloads what is required by the users. 
     65  * Cons 
     66   * Not upstream yet (still beta). 
     67   * A client that doesn't run the avahi server side doesn't share it's downloads (so all clients need to run the "server"). 
     68   * Requires working avahi on the network.