Bug #6821
closedMigrate to new MailPoet
0%
Description
The new MailPoet may fix of our performance issues from previous versions. But it has a migration routine that might not agree with our setup very well. Also, it appears to have some hardcoded restrictions on subscriber list size, which at least one of our existing users (pcp.gc.cuny.edu) will run afoul of.
Related issues
Updated by Boone Gorges about 8 years ago
- Related to Bug #6794: MailPoet subscriber click tracking causes excessive DB writes, leading to performance problems added
Updated by Matt Gold about 8 years ago
Please let me know when/if you think we should consider removing MailPoet as an available plugin for the Commons.
Updated by Boone Gorges about 8 years ago
If it were up to me, I'd make it unavailable immediately, at least for new sites. But it does seem to provide a valuable service to some sites, so it's worth investigating to make sure that we have alternatives in place.
Updated by Matt Gold about 8 years ago
Please do make it unavailable immediately for new sites. Thank you.
Updated by Matt Gold about 8 years ago
Honestly, I'd say that we should make it unavailable for everyone and write to users. If it is repeatedly bringing the Commons down and making the network as a whole unstable, as it appears to be, it is not a viable plug-in for the Commons IMO. Please let me know whether you agree.
Updated by Boone Gorges about 8 years ago
MailPoet can no longer be activated by non-super-admins: https://github.com/cuny-academic-commons/cac/commit/b2e1b4359a05378a90a00085f1533c14ff6ff376 It remains active on existing sites where it's already active.
Honestly, I'd say that we should make it unavailable for everyone and write to users. If it is repeatedly bringing the Commons down and making the network as a whole unstable, as it appears to be, it is not a viable plug-in for the Commons IMO. Please let me know whether you agree.
I've already sunk the time into working around the immediate problems. It may be worth developing a plan to get existing users off of the plugin, but I don't think it's urgent. A number of sites - PCP in particular, but also a couple others - are heavy users of the plugin, and at the very least will need to have some assistance getting their subscriber lists out of MailPoet. Assuming that we don't want to leave them in the cold, forcing a migration will create yet more work for the development and community teams. For the time being, as long as we don't experience more catastrophic failures, I'm in favor of leaving things as they are.
Updated by Matt Gold about 8 years ago
Okay -- that sounds like a good plan. Should the site instability become an issue again, though, let's reconsider. Thank you for all of your work on this over the past few weeks, Boone.
Updated by Boone Gorges over 7 years ago
- Status changed from New to Rejected
- Target version deleted (
1.11)
Given the amount of work sunk into this in the past, I'm going to table the migration process. If and when it becomes important to leave MailPoet 2.x, I'd prefer to spend the effort moving users to a different platform altogether.