Mercurial > dovecot > original-hg > dovecot-1.2
view TODO @ 1901:f59cd148bdfe HEAD
COPYING.MIT missing
author | Timo Sirainen <tss@iki.fi> |
---|---|
date | Thu, 25 Dec 2003 17:26:52 +0200 |
parents | 0b8260468474 |
children | 11c1f1203b65 |
line wrap: on
line source
- support for multiple auth checkers. passdb+userdb should be tied together. auth process configurations really should be used for this.. - openbsd: if we ever write() to a file that is mmaped(), even if munmap()ed in the middle, the mmaped area doesn't get updated. This mostly happens when updating headers with write().. is it worth fixing? .. - maildir append doesn't actually need the mailbox to be synced. index->open_mailbox() could have some parameter to specify if mailbox should be synced or not. - we could process multiple commands (especially command + implicit sync after it) without dropping (index) lock in the middle. that'd mean all commands should be able to begin in any lock state and they shouldn't drop the lock. there'd be some new drop_lock() function which finally does that. - imap_parser() should rather have a full-line memory limiting rather than per-parameter. some clients can send huge messageset parameters and it'd be nice to make the max. line length configurable - fix cygwin compile? _close and _read in *stream*.c fails. - mailbox_check_interval: we shouldn't send anything if write() returns EAGAIN. not all clients are reading input all the time and there's no point in sending updates when no-one is reading them. - mbox: we don't notice external flag changes if they don't change file size - Create new X-IMAPlog header which Dovecot uses internally to store changes. If change is in the log header, we don't have to reparse the whole file to find out what changed. Keep last 10 or so changes logged. First process that notices unknown change should figure out what changed and add it to log, to avoid everyone else doing that. - X-IMAPlog: <timestamp> <file offset>:<changed message count>, <file offset>:<changed message count> <timestamp2> ... - when writing the timestamp, make sure that we handle the situation when second just happens to change when we write the log header, and old timestamp is left into log. - when syncing or writing the file, keep the lock long enough that current second changes. this is to make sure that no-one else changes the file and make us miss the change because timestamp didn't change - recent counters are broken - the process that assigns UID to message should be the one to see it as recent - we don't handle non-contiguous recent ranges - don't break if we rename selected mailbox - struct mail_index_data_record_header isn't 8byte padded with 32bit off_t - handle out of quota conditions: - if dovecot-uidlist can't be written, assume the new mails have UIDs beginning from uidlist.next_uid. Whenever mails are expunged, overwrite the next_uid field with the current highest next_uid. Whenever we have assumed UIDs and uidlist gets updated, throw the client out with "inconsist mailbox". - make sure all syscalls check for ENOSPACE (and ENOACCESS while at it) - NFS safety: - use link()s instead of relying on O_EXCL - .subscriptions: use rename() like dovecot-uidlist - .customflags: use rename(), but there's a problem when we have to remove unused flags to make room for new ones. to fix that add new field in the file, it would be set for flags which are currently unused. if that field is set, .customflags must be locked before the flag is set to any messages. but make sure there's no race conditions, we probably have to wait a few seconds just to make sure no-one set a flag we want to remove - send client IP immediately after accept() to master process. make sure master shows the IP if login dies unexpectedly. master should probably also kill the login process if it doesn't kill itself soon enough.. or maybe just log the IP immediately. - .subscriptions and .customflags files conflict with personal namespace in maildir... and what about .imap.index* files if they get moved into root dir? - ioloop-kqueue.c patch - workaround: oe-nonsynced-uid-fetch (it's now done always) - does dovecot-auth really break when it runs out of fds? - add something to docs about locks in /var/mail - safe_mkdir_path() and create mbox / maildir paths - create auth_chroot dir at startup - remove Maildir/.INBOX/ - maildir: create foo/bar -> OK, delete foo/bar -> error - NOOP should complain about inconsistency immediately - and it complains about indexid changes next time.. - x LIST "" INBOX.% should contain "INBOX." in reply. But it breaks some clients, leave optional. - full_filesystem_access=yes + Maildir + SELECT "INBOX/" works? .. - should we support some non-tokenizing way to parse mail addresses? .. eg. "foo ? bar <x@y>" would now show up as "foo? bar <x@y>" - we hang if ssl key is password protected - DELETE/RENAME: when someone else had the mailbox open, we should disconnect it (when stat() fails with ENOENT while syncing). - bugs - SIGHUP didn't update imap_listen. this is a bit annoying to fix though, since new listen() may fail for a few times because login processes may not die immediately.. - SIGHUP doesn't update log file location. - CREATE a, CREATE b, save mails into them, DELETE a, RENAME b a. -> breaks if a+b have same UIDVALIDITY. We could update the UIDVALIDITY for the renamed mailbox and all mailboxes under it. Then return with tagged "OK [NEW-UIDVALIDITY 1234] Renamed". Assuming other IMAP people agree to that. - reliability fixes: - if we deleted mail from index but didn't write modify log, other dovecots don't handle it properly. they either assert at index-sync.c:42 or if new mails have also been added since, they don't notice it at all actually, that breaks reads as well since we get expunges only from the old file.. and check that deleting file does "inconsistency error" - if imap process notices that both modify logs are getting full because it's client isn't syncing, the client should be disconnected - we don't handle out of memory conditions too well, malloc failing kills the process which is good enough (and likely never happens), but mmap() failures aren't handled too well. Rather should be handled in similiar way to locking failures, so that at least we don't try to rebuild the index because of it. - limit folder hierarchy levels? user can now create eg. a/a/a/a/... and then start renaming them from end to beginning, which probably will at some point start causing syscall failures which will fill up logs. - fsck should check binary tree - dotlock overriding is racy, but it's pretty difficult to fix it. Also overriding someone else's dotlock in shared folder isn't possible. These could be fixed by having separate lock process running as root, which would chown() the file for another uid and then unlink() it as that user. One problem with that is that if malicious user sets setuid+execute bits on for the file, he could run the file and get changed to the new uid. That hopefully shouldn't matter much since the new uid should be user with minimum possible privileges. Anyway, optional.. - mbox - Move data within file instead of writing it to temp file. We can now do moving easily with o_stream_send_istream(). - if a file isn't valid mbox and it's tried to be opened, say it in one line in error log, not 6.. - When expunging the first message we could move the X-IMAPbase header to next message to avoid full rewriting later. - We shouldn't send X-IMAPbase, Status, X-Status, X-Keywords, X-UID, etc. headers to client - they may change and clients must see messages as immutable. Create istream-filter and use it. - two adjacent From-lines breaks us. not too easy to fix though. - Handle UW imapd's "DON'T DELETE THIS MESSAGE" message. - checks: - if we have entries in modifylog with UID 10..11, 9..12, 8..13 etc. do they work correctly? - make sure SELECT rebuilds index properly when next_uid is near 32bit value - make sure connection limits work - check that both header and envelope address writer produce same output - check if some asserts should be changed into if/i_panic code, so that disabling asserts would be possible - enhancements: - option to disable SORT, SEARCH and other memory/cpu-intensive features. defaults and per-user by dovecot-auth. - when fetching body/envelope/etc we could try to cache it immediately if we can get lock with try_lock. - maildir could support also the dirty-flag in messages. files would be renamed "whenever there's time" (that'd require the indexer program, or forking and doing it in background) - optionally keep the message file name as it's UID. Then we don't have to save the filename anywhere. - send EXISTS immediately after new mail arrives. - linux: we can use dnotify for maildir (but not mbox I think, we'd get interrupted all the time if we checked eg. large /var/spool/mail) - *bsd: kqueue() can notify changes in mbox and maildir - OpenSSL: support generated DH parameters - SSL: Support password protected key files. Support reading the password from user at runtime (dovecot startssl or something). - check with strace what dovecot does when evolution checks new mail, it's quite a lot. some things probably wouldn't need to be done (mkdirs/symlinking inbox) and other things could be cached in memory. - sort: we could create alternative binary tree file(s) for different sort conditions, ".tree-sort" or something. sort code itself already supports this optimization. - tree file: should we instead use b+-tree or something similiar? or at least try to do some defragmentation with it, so that the root nodes would be kept at the beginning of the file. - use vsftpd-like safebufs, ie. keep non-rwx page before and after the memory we want to use. - mmap_anon() - mmap()ing files would probably need to first go through anon_mmap() and then use MAP_FIXED. annoying that it slows the mmaping.. - data stack should use mmap_anon() - option: copy /var/mail/$user to INBOX when logged in. nice for not missing any mails with quota enabled - support zlib compressed mbox/maildir? mbox maybe just read-only. do it through istream-zlib wrapper - THREAD=ORDEREDSUBJECT - although pretty useless I'd think. - sql authentication and mail database support. create some generic lib-sql which both can use - maildir: change it to use '/' as hierarchy separator to allow '.' characters (for usernames in shared folders) - POP3: add support for LAST command, but make it optional since it has to play with \Seen flags (RSET would unset them all with it enabled, not very nice). - logging - Login: username 1.2.3.4:1025 5.6.7.8:993 imaps,compressed - Logout: username 1.2.3.4:1025 5.6.7.8:993 imaps,compressed in:1000 out:1000000 - n failed login attepts (before failure or success, once in n seconds) - lib-charset: - utf8_toupper() is a must. and a bit difficult if we want to do it right. - add support for other things than iconv() as well? we could reuse the code from cyrus or courier - cache iconvs? they'd probably be faster if we just reset the conversion instead of opening new one every time. and there will likely be only one or two charsets which are used for nearly all conversions. - should we allow following symlinks in mbox/maildirs? they are now. - if we implement shared mailboxes with shared indexes, never do that or others could symlink your personal mailboxes and see the indexes created for it which may contain envelope etc. data - this allows circular mailbox hierarchies which should be prevented by eg. allowing max. 20 hierarchies. - index: - currently we read-lock it when eg. fetching mails. that can be a very long lock. currently we also wait for exclusive lock if we want to sync mailbox. with shared folders this would be real problem, but it's not nice currently either if you want to have multiple connections to same mailbox. solutions: - syncing shouldn't wait lock more than a few seconds - fetching could drop the lock after a few seconds. might get tricky though. - maybe more fine grained locking? reading all mails shouldn't prevent appending new mails. accessing header could be more problematic. - we could try compressing same from/to/subject fields into a single location in data file. requires larger changes.. - Most messages are text/plain/7bit/us-ascii/no-other-content-type-params. Instead of saving tons of nearly identical BODY/BODYSTRUCTURE fields into cache, we could just set INDEX_MAIL_FLAG_TEXT_PLAIN_7BIT_ASCII bit on and generate the bodystructure for such messages on the fly. - read-only support for mailboxes where we don't have write-access - if we expunge last message in index, we could just decrease the used_file_size header instead of leaving hole into file - we should try to avoid completely rebuilding indexes unless they're corrupted. especially if we later want to support some read-only boxes and keep the mail flags only in index file. fsck() could verify that records are ok, and that if data file isn't ok the record is deleted. - if .customflags is removed and Maildir files have custom flags, add "unknown1" "unknown2" etc. flags to .customflags file for each found flag - skipping deleted records in .imap.index would be faster if we saved the deleted block size to first/last record, so we could just jump over them. - support storing message headers into indexes. this could be useful when indexes are in local disk but actual mails are accessed through NFS. - we could send flag changes after all commands by making expunge/flags sync counters separate for modify log. flags would need to update the seq though, too slow? - cost-based caching to indexes? when accessing the indexes, save how much extra data we had to read because we did/didn't cache some field, and also how much extra data we avoided reading because we had it cached. but is it fast enough to find out the lengths without uselessly slowing down? - if we wanted to support huge mailboxes with small memory usage, it'd now be possible if we just instead of mmap()ing the whole index files would have maybe 3-4 256k mmap()ed areas which we move based on the need. - should work fine with .imap.index and .imap.index.data - log files aren't affected by mailbox size - if the tree file also kept constantly moving the nodes so that tree's root was at the beginning of the file, we could use this mmap caching with it too - but, is it worth the trouble really? the OS can do all this itself, only thing we're doing is keeping the processes virtual memory usage small. - SEARCH: - message_body_search() could accept multiple search keywords so we wouldn't need to call it separately for each one (so we wouldn't need to parse the message multiple times). - message_body_search() could support NULL MessagePart and the searching could be done while parsing the message. this would need changes to message_parse() as well. - could optionally support scanning inside file attachments and use plugins to extract text out of them (word, excel, pdf, etc. etc.) - use a trie index for fast text searching, like cyrus squat? - Create our own extension: When searching with TEXT/BODY, return the message text surrounding the keywords just like web search engines do. like: SEARCH X-PRINT-MATCHES TEXT "hello" -> * SEARCH 1 "He said: Hello world!" 2 "Hello, I'm ...". This would be especially useful with the above attachment scanning. - general: - sieve (rfc3028), we can use Cyrus Sieve - rfc2231 continuation support - rfc2557 support for BODYSTRUCTURE, as specified by RFC3501 - lmtp server - is it needed? dovecot-deliver binary at least would be useful - create indexer binary - support Maildir++ quota - some kind of IMAP proxy for load distributing - maybe give more untagged NO/ALERT replies? like when mailbox is in inconsistent state. and when UIDs are reordered because they're too large. - imap/ and lib-imap/ should allow infinite number of custom flags, it's storage's problem if it can't handle too many of them. - things calling message_send() could verify that it wrote enough data. if not, fill the rest with spaces and return failure. -1 = error, 0 = filled, 1 = ok. - cleanups: - check if t_push()/t_pop() should be added somewhere - try to fix @UNSAFE code to use buffer API instead - subscription-file.c, custom_flags - [io]stream-file.c? - grep for FIXME - index/create_temp_file is used in only two places. once mbox rewriting doesn't need it, get rid of it. - auth / login: - kchuid, SRP, anonymous SASL - Digest-MD5: support integrity protection, and maybe crypting. Do it through login process like SSL is done? - for invalid user/pass, wait for a while before giving a reply to user - dovecot-auth should limit how fast authentication requests are allowed from login processes. especially if there's one login/connection the speed should be something like once/sec. also limit how fast to accept new connections. capabilities: - preferrably all should be possible to #ifdef away by a configure option (--without-capabilities=acl,namespace,...) - possibility to disable them from config file - acl (rfc2086, draft-ietf-imapext-acl), namespace (rfc2342) - probably do it like cyrus. "user.<username>" to access other users, with "" defaulting to "user.<myself>". these should be configurable however. - shared namespaces? maybe configurable in config file - easiest way to do ACL would be to use unix modes, but is that useful at all? Well, ACL2 has a bit better support for that, so maybe we could support it. - otherwise gets a bit trickly, we could keep all mail in "imapmail" group and 0600/0700 mode by default, but when mail is shared to others, the group read/write access bits would be set. or alternatively we could launch another imap process to handle it, which we should support anyway. ACLs could be stored into ".acl" ascii file in each folder. - support for private and shared flags, configurable by mailbox admin. this isn't in any draft yet, but ACL2 author was going to create one. [SHAREDFLAGS (...)] would specify which ones are shared, don't know yet how they would be configured. - quota (rfc2087, draft-cridland-imap-quota) - give filesystem values only to admins - support for Maildir++, probably no need to support more. quota capability supports complex quota configuration, but if no mailer supports them we probably shouldn't bother either - id (rfc2971) - must be configurable what gets sent, default to only name=Dovecot - separate pre/post-login settings - optionally log configured parts of the client information, but only once, probably at the same time as logging "Logged in", "Disconnected", etc. - remember to force truncating values longer than 30 chars, especially before logging - mailbox-referrals (rfc2193) - this is useful whenever we would otherwise need to make the connection ourself. for example load balancing and shared mailboxes requiring another UID to run. - this rfc defines no exact way for server to detect if client supports referrals or not. I don't think there's much point in supporting only referrals, as most clients don't support them. Instead we should return referrals when we know that client supports them, otherwise do the connecting ourself. If client issues RLIST or RLSUB command, it's safe to assume it supports referrals. - for load balancing this works just fine, but what about shared mailboxes which require different UID? If we login with our own username, we end up with our own UID instead of what we wanted. IMAP URLs don't support separated authorization id which would have made this very easy.. We could give the "userid@group" as userid, but clients probably treat it as different userid and ask the password again. - problems, problems, .. maybe not worth the trouble. - uidplus (rfc2359) - uid expunge: no problem - append, copy: oh no. these would slow down things and make handling them much more difficult. currently we just store the mails to destination mailbox without touching the indexes. since we'd need to know their final UID, we'd have to lock the indexes and mbox) fsck() first and append() next to find out the uid, maildir) move the mail directly into cur/ and index it. - except, the index updating is faster than requiring to sync the changed mailbox. we'll probably create such deliver-binary anyway, so it can't be that difficult to code. - drafts: - http://www.imc.org/ids.html - annotate (draft-ietf-imapext-annotate) - per-message annotations. this will be major change. especially because currently there's no suitable storage for them, and they'll probably change all the time.. maybe if we moved into berkeley db to store the .data file and these annotations. - this is separate problem from index files. indexes are treated as temporary files, annotations are permanent data. we'd have to support non-db way to do this too, which would probably be just a simple (slow) text file. - annotatemore (draft-daboo-imap-annotatemore) - server and per-mailbox annotations. much easier than per-message annotations, but they'd be easier to place into db as well. - binary (draft-nerenberg-imap-binary) - perhaps not too useful. I'd like to make Dovecot fully binary-safe though. - view (draft-ietf-imapext-view) - slow, complex, luckily draft expired almost two years ago. i hope i don't have to implement this :) - can be done client-side just fine (evolution's virtual folders)