Email search not working properly
-
As the OP I must say that I had to finally give up in enabling full text search as it kept breaking searches. It works for a while but then breaks again and people get no results for very simple searches. Rerunning the script manually to reindex every time is not really a solution.
-
@James nothing special about the mailbox other than it is the largest one at 1.12 GB. The total size of all mailboxes on the server is 2.72 GB and with 8GB RAM assigned to mail container, I don't think lack of memory is what is causing it to fail. This particular mailbox is the only one with an alias associated with it. I removed the alias, re-ran the rebuild-index.sh script but ended up with the same error. So that's not it either.
doveadm(mail4@domain.com): Error: fts_solr: Indexing failed: 500 Server Error doveadm(mail4@domain.com): Error: Mailbox INBOX: Transaction commit failed: FTS transaction commit failed: backend deinitI think I will follow OP's suit and disable the full text search until it is more stable on Cloudron. If you would like to have access to my server for some more troubleshooting, just let me know .
-
@James nothing special about the mailbox other than it is the largest one at 1.12 GB. The total size of all mailboxes on the server is 2.72 GB and with 8GB RAM assigned to mail container, I don't think lack of memory is what is causing it to fail. This particular mailbox is the only one with an alias associated with it. I removed the alias, re-ran the rebuild-index.sh script but ended up with the same error. So that's not it either.
doveadm(mail4@domain.com): Error: fts_solr: Indexing failed: 500 Server Error doveadm(mail4@domain.com): Error: Mailbox INBOX: Transaction commit failed: FTS transaction commit failed: backend deinitI think I will follow OP's suit and disable the full text search until it is more stable on Cloudron. If you would like to have access to my server for some more troubleshooting, just let me know .
Hello @hakunamatata
Yes, please write a mail to support@cloudron.io and reference this issue.
-
@James , I think I found the issue. After googling the time out error message, I came across some threads on other forums which seemed to indicate that it could be related to large email messages that are not being indexed within the 120000 ms / 2 Minutes timeout interval. So I went to mail4's INBOX , deleted the largest email (17.7MB) and then re-ran the script. This time it failed during indexing mail4's TRASH folder. Once I deleted the email from the TRASH folder, voila, indexing was successful.
So perhaps the timeout needs to be increased or be a user defined value.
In my case, I am running my Cloudron instance as a VM on a Synology DS1621+ which features an AMD Ryzen V1500B processor (4 cores, 8 threads, 2.2 Ghz base clock). It isn't a very fast processor so that could be a contributing factor.
-
@James , I think I found the issue. After googling the time out error message, I came across some threads on other forums which seemed to indicate that it could be related to large email messages that are not being indexed within the 120000 ms / 2 Minutes timeout interval. So I went to mail4's INBOX , deleted the largest email (17.7MB) and then re-ran the script. This time it failed during indexing mail4's TRASH folder. Once I deleted the email from the TRASH folder, voila, indexing was successful.
So perhaps the timeout needs to be increased or be a user defined value.
In my case, I am running my Cloudron instance as a VM on a Synology DS1621+ which features an AMD Ryzen V1500B processor (4 cores, 8 threads, 2.2 Ghz base clock). It isn't a very fast processor so that could be a contributing factor.
