Feature/vosa 284 support for backend bridge #1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- This parse and set the user defined configuration for dev image.
… image from the vm disk image.
- make: this will generated the vmdk image.
- ova : will create the ova file from the vmdk image (running vosa -i {instance} make first)
- Called parser from global scope instead of from a function to set the configuration properties globally.
… encrypted vmdk and ova file.
- Cause vosa does not use print_and_log command
- the call to get_publication_list didn't pass any port and the method expected one, falling back to the default if it wasn't provided. - deleted trailing white space.
- Fixed typo in ece-local-builder log message.
…erate man pages in package creation script.
- Before the first parameter was taken now we are using ${@} intead of $1.
- Renamed variable name from ova_config_eula ----> ova_config_agreement.
- Set default values for vendor, vendorurl and agreement.
- Memory, cpus and heap size need to be integer
…ication app_base.
- which is a good thing - should go into the standard webapps dir
- it's its own webservice now, see MENU-68
…d new maven cooridnates to download it
… connector - We often set up (as in, ece-install does this) two Engine/Service elements, one for each Engine, which again has its own HTTP connector to serve different webapps from different ports. If so (there's strictly speaking no need to have multiple Services to have multiple connectors, so we cannot assume this is always the case), the correct <Service/> element for the publication <Host/> elements is the second Service element, in other cases, the the first one. - FYI: Since both /Server/Service/Engine elements are the same XPATH wise, we have to access the correct Engine element by the index of the containing Service element.
…nce_is_running() - this method should ensure that the instance used for creating the publication is running, however, this method was broken, perhaps in ed37d3d, perhaps much longer ago. The symptom of this method failing was these strange messages in /var/log/escenic/ece-install.log: [ece#start-engine1] You must specificy a command, see 'ece help'. [ece#versions-engine1] You must specificy a command, see 'ece help'. - Improved the algorithm while I was at it. It now checks every second to see if ECE is bootstrapped, instead of just sleeping for a whopping 60 seconds regardless of the ECE running or not.
cue: backend_ece: http://otherhost:8080 Is used both for nginx/cors setup as well as the backends.yml configuration for CUE. Thus, it needs to get /webservice/index.xml dynamically appended. This has now been fixed.
- YAML equivalent:
profiles:
cue:
install: yes
backend_ece: http://ece.example.com;
backend_ece_local: http://localhost:8080;
backend_ng: http://ng.example.com;
- This _local setting is only needed if CUE and ECE run on the same
machine (and you use Live). If ECE runs on a separate machine, you
don't need the extra _local value.
- Longer explanation: If ECE runs on the same machine as CUE and
you've got CUE Live, you must specify backend_ece and this cannot be
a local reference since Live will use this value in the CUE plugin,
which runs in the browser and must thus be resolvable by the outside
world.
However, since ece-install will setup nginx forwarding using the
value of ECE's whereabouts, it must be able to differentiate between
the two addresses
- regression introduced in 9f4822a
"I just want to copy and paste a bunch of commands and send them to you"
- Improved introduction - Include all .conf and .yaml files in root's home directory → greater chances of hitting something useful 😉
- Debian/Ubuntu now (15.04+) uses systemd and this must respect fai_cache_port too
- not only edit and update them - added support for editing/updating/creating section feed publication resources. - example: $ ece --publication-resource section-feed --publication mypub create
This should fix the cases where upgrading e.g. escenic-content-engine-scripts will remove any local changes to /etc/default/ece by the user
Background: This allows /usr/share/escenic/escenc-video to be a link to a e..g /usr/share/escenic/escenic-video-323232 where /usr/share/escenic/escenic-video-323232 is a non-package installed.
- installs and configures the SSE proxy itself, it may have many
backends (typical ones for an Escenic architecture are ECE, Live and
Video).
- sets up nginx to proxy SSE proxy requests
- configures ECE to expose the changelog through the SSE proxy
- configures Live too
- added unit tests for the new configuration options
- The editor & presentation profiles now sets up an extra connector
for the SSE proxy
Configuration items:
-- fai_editor_install=1
-- fai_presentation_install=1
Will set up an extra Tomcat connector needed for the SSE proxy to
work. No configuration is necessary, it'll set up the default:
<Connector
port="8083"
protocol="HTTP/1.1"
connectionTimeout="20000"
URIEncoding="UTF-8"
compression="off"
redirectPort="8443"
proxyPort="80"
/>
- Example ece-install.yaml configuration:
profiles:
sse_proxy:
install: yes
exposed_host: proxy.quanah
backends:
- uri: http://localhost:8083/webservice/escenic/changelog/sse
user: foo_admin
password: foo
- Full list of options (equivalent entries to the older ece-install.conf
format):
fai_sse_proxy_exposed_host= # default is the machine name (which is
# probably not what you want), typically
# proxy.example.com
fai_sse_proxy_exposed_port= # optional, default is 80
fai_sse_proxy_install=
fai_sse_proxy_backends= # required
fai_sse_proxy_ece_port= # optional, used for tomcat conf
fai_sse_proxy_ece_redirect= # optional, used for editor & presentation
- updated reference doc
- SSE proxy can also be installed on RedHat
- also fixed an error message coming if there are no Escenic related man pages on the system. - none of these two errors caused any real problem, but they were false warnings/errors and were thus confusing to the user.
- Since some packages, with various logic, have been moved to to a cue sub directory (2017), we'll try again with 'cue' appended to the base URL
- fixes regression introduced in 740b16c
- support for 'service memcached' has been pulled (seems broken to be honest). - RedHat are firm believers in systemd now
modhassan
force-pushed
the
feature/VOSA-284-support-for-backend-bridge
branch
from
f63bcf2 to
f452f68
Compare
- An optional backend bridge option has been added to the CUE profile to
support the new bridge endpoint.
- If set, the generated `backend.yml` file gets an entry like:
cue:
backend:
bridge: http://bridge.example.com
modhassan
force-pushed
the
feature/VOSA-284-support-for-backend-bridge
branch
from
f452f68 to
24ce72f
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
New backend
backend_bridgeadded to support new endpointbridge