Bug 1312098

Hello Pradeep,

Based upon your testing, do you have any specific recommendations on the configurations/properties to tune and their values to support addressing this request?

Thanks!

The 6.1 docs linked in comment 0 are a bit out of date.  6.2 has different requirements, documented at  https://access.redhat.com/documentation/en-us/red_hat_satellite/6.2/html/installation_guide/large_deployment_considerations.

Each item is detailed below. Some are entirely possible in the installer.  The postgresql configuration is complicated, see item #3.  I would not expect that to make 6.3.


1. "Increasing the Maximum Number of File Descriptors for Apache"

Not available. Maybe expose in puppet-foreman? But installer won't wipe out these changes if manually made for now as changes are made via systemd drop-ins.


2. "Increasing the Maximum Number of File Descriptors for qpid"

   PR (pending) https://github.com/Katello/puppet-qpid/pull/64

   Custom hiera usage:

   ::qpid::open_file_limit: 65536

3. "Increasing the Shared Buffer and Work Memory" (postgresql)

   Complicated, because of how puppetlabs-postgresql works.  I wouldn't expect this to make the installer for 6.3

   Other BZ's: https://bugzilla.redhat.com/show_bug.cgi?id=1440879, https://bugzilla.redhat.com/show_bug.cgi?id=1449707,    https://bugzilla.redhat.com/show_bug.cgi?id=1421636

   Upstream discussion: https://groups.google.com/d/msg/foreman-dev/mKELbDvo-TQ/kYRIHYZQDAAJ


3. "Increasing Concurrent Content Host Registrations"

   Partial. See https://bugzilla.redhat.com/show_bug.cgi?id=1449697


4. "Increasing the maximum number of open files for qdrouterd"

   PR (pending) https://github.com/Katello/puppet-qpid/pull/50

   Custom hiera usage:

   ::qpid::router::open_file_limit: 3160

1) Tuned profiles

    If its RHEL6, install tuned and setup throuput-performance profile
    If its RHEL7, its taken care
2)  Apache Configuration : Passenger config
    It depends on workloads. 

3) Increase apache open files



4) Increase qpid open files

5) Increase content_action_accept_timeout
 in Administer -> Settings -> Katello
 to more than default (20seconds) so clients have more time to answer

6) Increase qdrouterd open files

7) increase postgresql tunings
   shared_buffer
   max_connections
8) httpd_foreman_ssl_katello_keep_alive: 'On'
9) httpd_foreman_ssl_katello_keep_alive_timeout: '5'
10) httpd_foreman_ssl_katello_max_keep_alive_requests: '0'
11) httpd_foreman_ssl_passenger_max_requests: '1000'
12) httpd_foreman_ssl_passenger_max_preloader_idle_time: '0'
13) httpd_passenger_passenger_max_pool_size

    This needs good capacity planning. increasing it more can cause more CPU usage. For puppet workloads also, it will affect

14) postgres_log_min_duration_statement: '500'
15) Increase smart_proxy open files

Based on Comment #5 moving this out of Sat 6.3 as its not possible.

2 PRS will be taken out of this 1 merged and another in the works see comment #5

*** Bug 1472587 has been marked as a duplicate of this bug. ***

Bug 1472587 has been marked as a duplicate of this bugzilla. 
 When addressing this one, please ensure to also cover the configuration parameters specified in: https://bugzilla.redhat.com/show_bug.cgi?id=1472587#c0

*** Bug 1244112 has been marked as a duplicate of this bug. ***

*** Bug 1440879 has been marked as a duplicate of this bug. ***

*** Bug 1440879 has been marked as a duplicate of this bug. ***

*** Bug 1326260 has been marked as a duplicate of this bug. ***

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2018:2927