Mercurial > dovecot > original-hg > dovecot-1.2
annotate TODO @ 9384:42f6578c8602 HEAD 1.2.5
Released v1.2.5.
author | Timo Sirainen <tss@iki.fi> |
---|---|
date | Sun, 13 Sep 2009 22:12:21 -0400 |
parents | 24952a7b371b |
children | 4e001a34078e |
rev | line source |
---|---|
9299 | 1 - acl: mail_debug=yes could log something useful |
9121 | 2 - dict quota: syncing may cause quota recalculation, which in turn syncs |
3 all mailboxes and then we'll assert-crash to avoid infinite looping | |
4 | |
9179 | 5 - acl: users are never dropped from acl_shared_dict. |
9056 | 6 - index-sync.c: line 39 (index_mailbox_set_recent_uid): assertion failed: (seq_range_exists(&ibox->recent_flags, uid)) |
7 ^ when stress testing and around "Duplicate file entry" errors | |
8831 | 8 - convert plugin: convert_pop3_uidl_format setting? so old %f uidls could be |
9 converted to dbox.. | |
8564
3c4934783aff
Mention SSL/TLS in "plaintext auth disallowed" error.
Timo Sirainen <tss@iki.fi>
parents:
8555
diff
changeset
|
10 - proxying: support fallbacking to local (or other?) server if the first |
3c4934783aff
Mention SSL/TLS in "plaintext auth disallowed" error.
Timo Sirainen <tss@iki.fi>
parents:
8555
diff
changeset
|
11 one is down |
8555 | 12 - fts-solr: handle DELETE, RENAME |
13 - fsck -> log_file_tail_offset 2273345664 -> 996 -> | |
14 mail-transaction-log.c: line 341 (mail_transaction_log_set_mailbox_sync_pos): | |
15 assertion failed: (file_offset >= log->head->saved_tail_offset) | |
8488 | 16 - virtual: "Searched n% of the mailbox" gives broken numbers since |
8555 | 17 ctx->seq jumps around. And why is it also returned when fts is enabled |
18 along with "Indexed n% of the mailbox"? | |
8488 | 19 - how do shared mailboxes work with plugins? |
20 - expire: not too well. would require knowing the mapping between shared | |
21 namespace and the original user's namespace to avoid duplication. | |
22 - lazy-expunge, fts, etc.? | |
23 - dovecot-acl-list: | |
24 - how does it work with global acls? | |
25 - update immediately after SETACL: add/remove entries, update timestamps | |
26 - read the entire file to memory only once and keep it there, stat() later | |
27 to see if it has changed. if not, perhaps don't even bother stat()ing | |
28 dovecot-acl files? at least not that often.. | |
9299 | 29 - when reading dovecot-acl file and seeing a +l right on a mailbox not |
30 listed in dovecot-acl-list, recreate it | |
8488 | 31 - add anonymous environment for anon logins |
32 - fs quota: getquotaroot inbox vs. other-box should return different quotas | |
33 if two quotas are defined | |
34 - deliver: log mailbox name using utf8, not mutf7 | |
8158 | 35 - new primes code: are hash tables now being resized too often? |
36 - auth_log_prefix setting similar to mail_log_prefix | |
37 - LDAP attrs: uid=foo,uid=bar doesn't work | |
38 | |
8488 | 39 - easily limit master users to be able to only log as other users within |
40 their domain | |
8158 | 41 - thread indexes: if we expunge a duplicate message-id: and we have a sibling |
42 with identical message-id:, we can probably just move the children? | |
43 (unless there are non-sibling duplicates) | |
44 - SEARCH INTHREAD requires no thread sorting by date - don't do it | |
45 - CONDSTORE: use per-flag/per-keyword conflict checking | |
46 - QRESYNC: Drop expunges from the middle of given seq sets if possible | |
47 - use universal hash functions? | |
48 | |
8488 | 49 - expire plugin: log more with mail_debug=yes |
50 - expire-tool -v could log UID and expire timestamps and what messages got | |
51 expunged | |
8247 | 52 - fts-squat: support ORs |
8158 | 53 - UIDVALIDITY changed while saving -> sync errors |
54 - mbox: copy to Trash, manually delete copied msg, change uidvalidity, | |
55 set nextuid=1, copy again -> error | |
56 - recent_uids assert at least with mbox | |
57 - quota fs: Should values returned by quota be divided by the actual | |
58 filesystem block size instead of hardcoded DEV_BSIZE? not with AIX.. | |
59 - mailbox list fs: Listing subscriptions with children return options doesn't | |
60 work unless iter_is_mailbox() returns the children flags | |
7502 | 61 - sieve-cmu.c crash: i_assert(buf->used - 1 == part->body_size.physical_size); |
7449 | 62 - convert plugin: Create a r/w lock for a file. It's read-locked if |
63 conversion isn't wanted and released when process dies. If conversion is | |
64 wanted and write-lock succeeds, conversion is done, if write-lock doesn't | |
65 succeed it fallbacks to using the old storage. When process is exiting it | |
66 again tries to write-lock and do the conversion. Add a parameter that | |
67 specifies if conversion should be done. | |
7259 | 68 - lucene: handle replacement chars? |
69 - squat: | |
7380 | 70 - wrong indexid |
71 - fts_build_init() assertion failed: (last_uid < last_uid_locked) | |
7259 | 72 - is locking done right? it reads header without file being locked? |
73 - split after ~8 bytes? | |
7391 | 74 - expunges are delayed until more mails are added |
7259 | 75 - test replacement chars (SEARCH / SORT / Squat) |
6987 | 76 |
7575 | 77 - dbox: |
78 - "File unexpectedly lost" doesn't get fixed by itself | |
79 - Fix support for multi-message files | |
7603 | 80 - Delete dovecot-keywords and dovecot-uidlist after all maildir files |
81 have been converted to native dbox | |
7449 | 82 - DEBUG: buffer overflow checking code probably doesn't handle a successful |
83 t_try_realloc() or pool_alloconly_realloc() properly | |
6987 | 84 - cache: compress when we can drop temporary fields. |
85 - new %modifier for reverse DNS lookups with a cache | |
86 - auth_gssapi_hostname = %Xl | |
87 - proxying would also want DNS lookups, but not reverse.. | |
8738 | 88 - ldap: |
89 - domain lookups which set the base for user lookup | |
90 - same attribute can't be used for multiple values. | |
91 - multiple attributes can't be merged to a single value. | |
92 - implement something like: | |
93 user_attrs { | |
94 uid = %{ldap:uidNumber} | |
95 home = %{ldap:homeDirectory} | |
96 quota_bytes = *:bytes=%{ldap:quota} | |
97 } | |
5621 | 98 |
7575 | 99 - Per-user options: |
100 - Deny deleting non-empty mailboxes | |
101 - Disable IDLE "still here" notifications | |
102 | |
6105 | 103 - maildir+pop3/deliver fast updates: |
104 - with locking enabled, pop3 could just keep the one and same sync lock and | |
105 do the whole thing using sync transaction | |
106 - don't update dovecot-uidlist if dovecot.index.cache doesn't exist / | |
107 there's nothing to cache | |
108 - if all messages are expunged and there are no unknown extensions in index, | |
109 unlink dovecot.index and rotate log and add some initial useful info to | |
110 the log (uidvalidity, nextuid) | |
5621 | 111 |
6105 | 112 - maildir |
6987 | 113 - don't allow more than 26 keywords |
6105 | 114 - physical separator could be configurable |
115 - maildir_copy_with_hardlinks: We're currently first hardlinking to tmp/ and | |
116 then rename()ing. This wouldn't be necessary if uidlist syncing noticed | |
117 that someone else already had added them to uidlist, and the existing UIDs | |
118 could be assigned to them in the index. | |
6754 | 119 - copying should copy already-cached data |
6105 | 120 - maildir_copy_preserve_filename=yes has a race condition causing "Append with |
121 UID n, but next_uid = y" errors when quota plugin is loaded. Practically | |
122 won't happen except in stress testing. | |
5621 | 123 |
6105 | 124 - mbox |
125 - UID renumbering doesn't really work after all? | |
126 - still problems with CRLF mboxes.. | |
127 | |
6987 | 128 - proxy: If remote server disconnects on login: |
129 login: tried to change state 2 -> 2 | |
6105 | 130 - logging consistency: |
131 http://www.dovecot.org/list/dovecot/2007-April/021532.html | |
132 - EXPUNGE command in read-only mailbox should give an error message if | |
133 there are messages marked as \Deleted? | |
134 - dovecot -o setting=something overriding | |
5557 | 135 - file_cache: we're growing the mmap in page size blocks, which is horribly |
136 slow if mremap() doesn't exist. | |
137 - login_max_processes_count shouldn't count proxying processes | |
138 | |
139 - Allow %variables in mail_chroot setting | |
140 - ssl_verify_client_cert isn't working if the SSL cert doesn't have CRL | |
141 | |
3087 | 142 - keywords: |
143 - add some limits to how many there can be | |
144 - don't return \* in PERMANENTFLAGS when we're full | |
145 - remove unused keywords? | |
2774 | 146 |
147 - caching | |
3904 | 148 - force bits should be used only for nonregistered fields |
149 - change envelope parsing not to use get_headers() so imap.envelope can | |
150 actually be cached without all the headers.. | |
6105 | 151 - if there's no other pressure for compression, we should do it when |
152 enough temp fields are ready to be dropped | |
3904 | 153 - we could try compressing same field values into a single |
154 location in cache file. | |
7259 | 155 - place some maximum limit of fields to cache file? maybe some soft and |
156 hard limits, so when soft limit is reached drop fields that have | |
157 been used only once. when hard limit is reached drop any fields to get | |
158 more space. all this to avoid cache file growing infinitely. | |
2774 | 159 |
3087 | 160 - mbox |
3904 | 161 - syncing existing indexes takes 4x longer than creating new one, why? |
162 - how well does dirty sync + status work? it reads the last mail every | |
163 time? not very good.. | |
164 - always add empty line. make the parser require it too? syncing should | |
165 make sure there always exists two LFs at end of file. raw-mbox-stream | |
166 should make sure the last message ends with LF even if it doesn't exist | |
167 in the file | |
168 - Quote "From ", unquote ">From " | |
4449 | 169 - COPY doesn't work to itself (lock assert crash, for now just disallowed) |
2143 | 170 |
171 - index | |
3087 | 172 - read-only support for mailboxes where we don't have write-access |
6105 | 173 - index file format changes: |
174 - pack UIDs to beginning of file with UID ranges | |
175 - use squat-like compressed uid ranges everywhere | |
6782 | 176 - write first extension intros in dovecot.index.log always with names |
177 - or better yet, drop the intro concept completely as it is now | |
178 - add "transaction boundaries" so we know which records belong to a | |
179 single transaction. | |
180 - only after that we can remove the transaction log offset | |
181 overwriting (otherwise we can get partial transactions in views) | |
2143 | 182 |
4449 | 183 - namespaces |
184 - namespaces: add new "auto_disable" flag so if the mailbox can't be opened | |
185 (eg. file doesn't exist), just ignore the problem and disable the | |
186 namespace | |
187 | |
188 - lib-storage | |
5557 | 189 - rename: allow moving between storages, as long as they're of same type |
4449 | 190 - x search charset asdf all -> should fail |
3087 | 191 |
192 - login | |
4449 | 193 - imap-login: Master sent reply with unknown tag 1. client closed |
194 connection at the exact same time master was logging it in? | |
195 see master_request_abort() | |
3087 | 196 - Digest-MD5: support integrity protection, and maybe crypting. Do it |
197 through login process like SSL is done? | |
198 - x login foo bar | |
199 x NO Authentication failed. | |
200 x login cras pass | |
201 * BYE Disconnected for inactivity. | |
202 ^ but it's not disconnecting! (buggy dovecot-auth not replying) | |
3744 | 203 probably because userdb lookup didn't reply, and fd was already sent |
204 for master.. should imap-login be handling it anymore?.. | |
3087 | 205 - imap-login: Authenticate PLAIN failed: Authentication failed: |
206 Authentication server isn't connected, try again later.. [127.0.0.1] | |
207 ^ NO Authentication failed. (should be Temporary login failure!) | |
208 - if auth process dies, login process should retry authentication if | |
209 possible. or if not, disconnect the client so it doesn't think the auth | |
210 failed. | |
474 | 211 |
3087 | 212 - auth |
4449 | 213 - with blocking passdb we're not caching lookups if the password was wrong |
5557 | 214 - non-plaintext authentication doesn't support all features: |
215 - multiple passdbs don't work, only the first one is used | |
216 - auth cache's last_success password change check doesn't exist | |
9121 | 217 - auth_cache_negative_ttl doesn't check password mismatches |
3087 | 218 - SIGHUP restarts auth processes .. but does it wait until they've finished |
219 with all requests? no. | |
220 - does dovecot-auth really break when it runs out of fds? | |
4449 | 221 - dovecot-auth should limit how fast authentication requests are allowed |
222 from login processes. especially if there's one login/connection the speed | |
3087 | 223 should be something like once/sec. also limit how fast to accept new |
224 connections. | |
225 - support read-only logins. user could with alternative password get only | |
226 read-access to mails so mails could be read relatively safely with | |
227 untrusted computers. Maybe always send [ALERT] about the previous | |
228 read-only login time with IP? | |
4449 | 229 - dovecot-auth workers: create a separate dovecot-pam worker which shares |
230 pretty much all code with dovecot-auth worker but isn't linked against | |
231 any libraries. or..? this might be difficult to do, especially because the | |
232 workers currently can handle any kind of passdb/userb requests. perhaps | |
233 there should be a completely separate simple PAM authenticator binary. | |
526 | 234 |
3087 | 235 - master |
236 - configurable syslog prefix | |
3450 | 237 - if there are duplicate settings, complain about it |
3087 | 238 |
239 - quota | |
240 - if dovecot-uidlist can't be written, assume the new mails have UIDs | |
241 beginning from uidlist.next_uid. Whenever mails are expunged, overwrite | |
242 the next_uid field with the current highest next_uid. Whenever we have | |
243 assumed UIDs and uidlist gets updated, throw the client out with | |
244 "inconsist mailbox". | |
245 | |
246 - ssl | |
247 - add setting: ssl_options = bitmask. by default we enable all openssl | |
248 workarounds, this could be used to disable some of them | |
249 - gnutls support isn't working | |
965 | 250 |
3087 | 251 - search |
6232 | 252 - message header search: we should ignore LWSP between two MIME blocks(?) |
253 - message_search_init() could accept multiple search keywords so we | |
1250 | 254 wouldn't need to call it separately for each one (so we wouldn't need |
255 to parse the message multiple times). | |
256 - could optionally support scanning inside file attachments and use | |
257 plugins to extract text out of them (word, excel, pdf, etc. etc.) | |
258 - Create our own extension: When searching with TEXT/BODY, return | |
259 the message text surrounding the keywords just like web search engines | |
260 do. like: SEARCH X-PRINT-MATCHES TEXT "hello" -> * SEARCH 1 "He said: | |
261 Hello world!" 2 "Hello, I'm ...". This would be especially useful with | |
262 the above attachment scanning. | |
0 | 263 |
3087 | 264 - general |
4449 | 265 - stop using atol(), atoi(), strtoul() etc. in places where we actually |
266 care about what they return, and rather create our own function which | |
267 checks if the input overflows the integer, and if so call i_fatal() | |
268 - LMTP server | |
269 - ability to build plugins statically into the binaries | |
3087 | 270 - ~/.dovecotrc to override system wide settings. namespace settings should |
271 override all the previous namespace settings instead of adding new. | |
272 - things break if next_uid gets to 2^32 | |
376
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
273 |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
274 capabilities: |
3087 | 275 - THREAD=ORDEREDSUBJECT - although pretty useless I'd think. |
376
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
276 - mailbox-referrals (rfc2193) |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
277 - this is useful whenever we would otherwise need to make the |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
278 connection ourself. for example load balancing and shared mailboxes |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
279 requiring another UID to run. |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
280 - this rfc defines no exact way for server to detect if client |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
281 supports referrals or not. I don't think there's much point in |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
282 supporting only referrals, as most clients don't support them. |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
283 Instead we should return referrals when we know that client |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
284 supports them, otherwise do the connecting ourself. If client |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
285 issues RLIST or RLSUB command, it's safe to assume it supports |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
286 referrals. |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
287 - for load balancing this works just fine, but what about shared |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
288 mailboxes which require different UID? If we login with our own |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
289 username, we end up with our own UID instead of what we wanted. |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
290 IMAP URLs don't support separated authorization id which would |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
291 have made this very easy.. We could give the "userid@group" as |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
292 userid, but clients probably treat it as different userid and |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
293 ask the password again. |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
294 - problems, problems, .. maybe not worth the trouble. |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
295 - drafts: |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
296 - http://www.imc.org/ids.html |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
297 - annotate (draft-ietf-imapext-annotate) |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
298 - per-message annotations. this will be major change. especially |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
299 because currently there's no suitable storage for them, and |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
300 they'll probably change all the time.. maybe if we moved into |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
301 berkeley db to store the .data file and these annotations. |
1178 | 302 - this is separate problem from index files. indexes are treated as |
303 temporary files, annotations are permanent data. we'd have to | |
304 support non-db way to do this too, which would probably be just a | |
305 simple (slow) text file. | |
5621 | 306 - use lib-dict probably |
307 - metadata (draft-daboo-imap-annotatemore) | |
376
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
308 - server and per-mailbox annotations. much easier than |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
309 per-message annotations, but they'd be easier to place into |
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
310 db as well. |
5621 | 311 - lib-dict, again |
376
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
312 - binary (draft-nerenberg-imap-binary) |