-
Notifications
You must be signed in to change notification settings - Fork 488
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
This is a Mother Issue for all the installer-related fixes and improvements addressed in 4.3 #3016
Comments
OK, sleep was addressed need to verify: |
api endpoints automatically blocked/ policy in place. |
confirmed can run installer as user that is not root. Closing. |
Just adding this to the issue, for posterity. Issues addressed:
Didn't do anything about:
(these could be addressed in another release) |
- Keep the list of default values up to date. - Consistent capitalization of PostgreSQL and Solr.
The "child issues" addressed are:
#2980
#1455
#1201
#1167
#1097
On top of these, the following has been addressed:
Removed the annoying long "sleep" statements; the installer should be taking a couple of minutes less time to run now.
It now takes care of the Jhove configuration file and schema location.
Improved the messaging and diagnostics.
Some large chunks of unnecessary output is now being hidden from the user (for example - the multiple screenfulls of output from the metadata block ingests).
Worked with Phil to address some security issues, I believe we have finally closed some particularly bad security holes that we used to leave open out of the box.
Updated accordingly and somewhat improved the installation guide.
(there are probably more small things I can't remember right now).
Most importantly, rewrote the installer so that it doesn't require running it as root.
Overall, it doesn't necessarily make the install process easier for the user. It's a bit of a "you get some, you lose some". The prerequisites section on Postgres got simplified a little bit; since it's no longer necessary to separately configure access on the local sockets connections... But then there is a bit of extra work - since now the user needs to supply the Postgres admin password, or set the access level to "trust" on localhost, in order to allow the installer to create the database and user...
It has however simplified the script internally and made it more robust. There are now fewer opportunities for things to go wrong. Switching between root and regular user there was occasionally causing permission errors and security warnings.
QAing the installer is notoriously difficult. I'm almost positive some user with some unusual setup will be able to break it. I'm willing to even maybe give my direct email in the release notes, telling people to contact me if they are running into problems with the installer.
But even if they have to send a support request through normal channels, installer script fixes are usually pretty easy. Unlike bugs in the app, we don't have to tell them to wait till the next release - usually I can just change a couple of lines in the script and give it back to them, here, try this again.
The text was updated successfully, but these errors were encountered: