Windows 10 will reach EOS (end of support) on October 14, 2025. For more information, see this article.

Search Support

Avoid support scams. We will never ask you to call or text a phone number or share personal information. Please report suspicious activity using the “Report Abuse” option.

Learn More

Thunderbird as thing email (IMAP) client with working search (gloda) - how to? (or feature request)

  • 4 replies
  • 0 have this problem
  • 1 view
  • Last reply by k.gajda

more options

Hey

Thunderbird has a well-functioning Search function that uses the cache (or rather, what the background search process/function "read" and saved in global-messages-db.sqlite).

With a newly configured account (IMAP), it works like this:

- establishes the directory tree - downloads message titles (including sender, date, etc.) - *.msf summary files - then a process that downloads the entire message for local storage - in the background, an indexing process (GLODA?) starts, which creates a search database (subject, from, to, keywords in the message body) using the global-messages-db.sqlite file.

For very large mailboxes (>100k msg, ~50GB), this means that the server's content is transferred to the local disk (and stored in ImapMail).

The question is: how do I configure TB to run in "thin" mode, meaning after connecting a new account: - establish a directory tree - download message titles (including sender, date, etc.) -> *.msf - DO NOT download messages for local storage - start the indexing process, which will download and index the subject, from, to, and keywords in the message body (it will download and "read" the message, but not save it locally) and save it to global-messages-db.sqlite

Setting "mail.server.default.offline_download" to "false" will cause this behavior, but the message body will not be indexed (only from, to, and subject).

I tried to work around this by: - configuring the mailbox with the download option for offline use - waiting for it to download - setting "mail.server.default.offline_download" to "false" - deleting all "cur" directories (I use maildir) or deleting all *.eml files from the ImapMail directory.

Content search works for messages that have been downloaded (and indexed) previously; it doesn't work for new messages.

Why am I trying this? Some people have "air" laptops (meaning - weaker processors, small SSDs) and need access to email (via IMAP), including full search, without having all their email downloaded locally.

In other words, how can I configure TB via IMAP to: - download the directory and subdirectory tree - download the headers of all messages - "read" and index all messages (including their content) and store them in "global-messages-db.sqlite," but not store all messages locally (online mode only)?

Regards Chris

Hey Thunderbird has a well-functioning Search function that uses the cache (or rather, what the background search process/function "read" and saved in global-messages-db.sqlite). With a newly configured account (IMAP), it works like this: - establishes the directory tree - downloads message titles (including sender, date, etc.) - *.msf summary files - then a process that downloads the entire message for local storage - in the background, an indexing process (GLODA?) starts, which creates a search database (subject, from, to, keywords in the message body) using the global-messages-db.sqlite file. For very large mailboxes (>100k msg, ~50GB), this means that the server's content is transferred to the local disk (and stored in ImapMail). The question is: how do I configure TB to run in "thin" mode, meaning after connecting a new account: - establish a directory tree - download message titles (including sender, date, etc.) -> *.msf - DO NOT download messages for local storage - start the indexing process, which will download and index the subject, from, to, and keywords in the message body (it will download and "read" the message, but not save it locally) and save it to global-messages-db.sqlite Setting "mail.server.default.offline_download" to "false" will cause this behavior, but the message body will not be indexed (only from, to, and subject). I tried to work around this by: - configuring the mailbox with the download option for offline use - waiting for it to download - setting "mail.server.default.offline_download" to "false" - deleting all "cur" directories (I use maildir) or deleting all *.eml files from the ImapMail directory. Content search works for messages that have been downloaded (and indexed) previously; it doesn't work for new messages. Why am I trying this? Some people have "air" laptops (meaning - weaker processors, small SSDs) and need access to email (via IMAP), including full search, without having all their email downloaded locally. In other words, how can I configure TB via IMAP to: - download the directory and subdirectory tree - download the headers of all messages - "read" and index all messages (including their content) and store them in "global-messages-db.sqlite," but not store all messages locally (online mode only)? Regards Chris

All Replies (4)

more options

You are asking for a solution that does not download messages to local storage, but also downloads for full indexing. The act of downloading all messages will, in itself, require resources. For a thin execution, the default is to just download headers until a specific message is selected. The headers are available for indexing, but only from header content. Indexing body content requires that the message be downloaded. That's how IMAP works. This URL may be useful: https://support.mozilla.org/en-US/kb/imap-synchronization

more options

Ok ok. Let the TB to download message temporarily (for indexing purposes) then delete it (from local tmp) after process is complete.

The "thin" instance of TB should work similarly to a web browser – it downloads information (text/graphics) to a cache for presentation purposes (website/web application), but this cache can be deleted. In the meantime, a history of your browsing activity is created, even though the cache has already been deleted.

In other words - one-time full synchronization with "feeding" the search in TB + successive "feeding" the search with new emails, without storing a local copy permanently (it is stored only, for example, until the program is closed when the user has read a given email)

more options

You're welcome to submit the idea, but I have doubts. Writing new code to download an IMAP account, then index it, and then delete the messages seems a horror show and high resource usage, the opposite of what you want. Good luck. You can submit suggestions to connect.mozilla.org and select the IDEAS section. This is not the place to accept suggestions.

Ask a question

You must log in to your account to reply to posts. Please start a new question, if you do not have an account yet.