Hub/Switch Installation Guide

Chapter 2 HPSS Planning
142 September 2002 HPSS Installation Guide
Release 4.5, Revision 2
of time. This makes a ‘migrate early, migrate often’ strategy a feasible way to keep XFS disks clear
of inactive data.
The only inherent size limitation for XFS is a 2 TB maximum filesystem size, which is a limitation
of the Linux kernel. The file size limit is 16-64TB (depending on page size) which is limited in
practice by the maximum filesystem size.
2.11.15 Gatekeeping
Sitesmaychoose toimplementsite policyinthe GatekeeperServerforload balancingcreate, open,
and/or stage requests. The site policy could limit the maximum number of non-AuthorizedCaller
requests allowed at once by either delaying or denying particular requests. To delay the request,
thesite policymayreturnaspecial retrystatus alongwith thenumberof secondstowait beforethe
Client API retries the request. Delaying requests should limit the number of create, open, and/or
stage requests performed at a particular point in time, thus decreasing the load on the system.
However, care must be taken to figure out the best retry wait scheme to meet the requirements for
each site and to configure the correct number of Gatekeeper Servers if the load on one Gatekeeper
Serverisheavy.(Note:ThemaximumnumberofGatekeeperServersperstoragesubsystemisone.)
Also, sites need to write their Site Interfaces optimally to return in a timely manner.
Twospecialerrorstatuscodes(HPSS_ETHRESHOLD_DENYandHPSS_EUSER_DENY)maybe
used to refine how a site may deny a create, open, or stage requests. If the BFS receives either of
these errors, then it will return this error directly to the Client API rather than performing a retry.
Errorsotherthan thesetwo orthe specialHPSS_ERETRYstatuswillberetriedseveral timesby the
BFS. See either volume of the HPSS Programmer's Reference for more information.
Create, open, and stage requests from Authorized Callers (DMG, MPS, NFS) can NOT be delayed
or denied due to timing sensitivity of the special requests these servers make to the BFS. For
example, migration of a file by MPS is an Authorized Caller Open request. The site policy could
keep track of Authorized Caller requests to further limit non-AuthorizedCaller requests.
If a Gatekeeper Server is being used for Gatekeeping Services, then the BFS for each storage
subsystem configured to use a particular Gatekeeper Server will return errors for the create, open,
and/or stage requests being monitored by that Gatekeeper Server when that Gatekeeper Server is
down. For example, if storage subsystem #2 is configured to use Gatekeeper Server #2, and
Gatekeeper Server #2 is monitoring open requests and is DOWN, then each open by the BFS in
storage subsystem #2 will eventually fail after retrying several times.
2.12 HPSS Metadata Backup Considerations
This Section contains guidelines for proper maintenance of the SFS metadata. The policies
described should be fully understood and implemented to protect the HPSS metadata. Failure to
follow these policies can lead to unrecoverable data loss.
Also, if running a Highly Available HPSS, make sure to review Section G.7: Metadata Backup
Considerations on page 569 as it covers issues related to reliable metadata backup in an HA environment.
Theremainderofthissectionisasetof“rules”associatedwith backingupHPSSmetadata.Though
toolslikesfsbackup aswellas sitespecificgeneratedscripts and procedurescan beusedtobackup