Mercurial > dovecot > original-hg > dovecot-1.2
annotate TODO @ 9593:24a891dfde36 HEAD
Added tag 1.2.13 for changeset b386f997c5c5
| author | Timo Sirainen <tss@iki.fi> |
|---|---|
| date | Fri, 23 Jul 2010 21:56:11 +0100 |
| parents | fa4b9d520687 |
| children |
| rev | line source |
|---|---|
| 9471 | 1 - antispam plugin: deleting mails from spam mailbox causes |
| 2 "dovecot.index reset, view is now inconsistent" | |
| 9299 | 3 - acl: mail_debug=yes could log something useful |
| 9121 | 4 - dict quota: syncing may cause quota recalculation, which in turn syncs |
| 5 all mailboxes and then we'll assert-crash to avoid infinite looping | |
| 6 | |
| 9179 | 7 - acl: users are never dropped from acl_shared_dict. |
| 9056 | 8 - index-sync.c: line 39 (index_mailbox_set_recent_uid): assertion failed: (seq_range_exists(&ibox->recent_flags, uid)) |
| 9 ^ when stress testing and around "Duplicate file entry" errors | |
| 8831 | 10 - convert plugin: convert_pop3_uidl_format setting? so old %f uidls could be |
| 11 converted to dbox.. | |
|
8564
3c4934783aff
Mention SSL/TLS in "plaintext auth disallowed" error.
Timo Sirainen <tss@iki.fi>
parents:
8555
diff
changeset
|
12 - 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
|
13 one is down |
| 8555 | 14 - fts-solr: handle DELETE, RENAME |
| 15 - fsck -> log_file_tail_offset 2273345664 -> 996 -> | |
| 16 mail-transaction-log.c: line 341 (mail_transaction_log_set_mailbox_sync_pos): | |
| 17 assertion failed: (file_offset >= log->head->saved_tail_offset) | |
| 8488 | 18 - virtual: "Searched n% of the mailbox" gives broken numbers since |
| 8555 | 19 ctx->seq jumps around. And why is it also returned when fts is enabled |
| 20 along with "Indexed n% of the mailbox"? | |
| 8488 | 21 - how do shared mailboxes work with plugins? |
| 22 - expire: not too well. would require knowing the mapping between shared | |
| 23 namespace and the original user's namespace to avoid duplication. | |
| 24 - lazy-expunge, fts, etc.? | |
| 25 - dovecot-acl-list: | |
| 26 - how does it work with global acls? | |
| 27 - update immediately after SETACL: add/remove entries, update timestamps | |
| 28 - read the entire file to memory only once and keep it there, stat() later | |
| 29 to see if it has changed. if not, perhaps don't even bother stat()ing | |
| 30 dovecot-acl files? at least not that often.. | |
| 9299 | 31 - when reading dovecot-acl file and seeing a +l right on a mailbox not |
| 32 listed in dovecot-acl-list, recreate it | |
| 8488 | 33 - add anonymous environment for anon logins |
| 34 - fs quota: getquotaroot inbox vs. other-box should return different quotas | |
| 35 if two quotas are defined | |
| 36 - deliver: log mailbox name using utf8, not mutf7 | |
| 8158 | 37 - new primes code: are hash tables now being resized too often? |
| 38 - auth_log_prefix setting similar to mail_log_prefix | |
| 39 - LDAP attrs: uid=foo,uid=bar doesn't work | |
| 40 | |
| 8488 | 41 - easily limit master users to be able to only log as other users within |
| 42 their domain | |
| 8158 | 43 - thread indexes: if we expunge a duplicate message-id: and we have a sibling |
| 44 with identical message-id:, we can probably just move the children? | |
| 45 (unless there are non-sibling duplicates) | |
| 46 - SEARCH INTHREAD requires no thread sorting by date - don't do it | |
| 47 - CONDSTORE: use per-flag/per-keyword conflict checking | |
| 48 - QRESYNC: Drop expunges from the middle of given seq sets if possible | |
| 49 - use universal hash functions? | |
| 50 | |
| 8488 | 51 - expire plugin: log more with mail_debug=yes |
| 52 - expire-tool -v could log UID and expire timestamps and what messages got | |
| 53 expunged | |
| 8247 | 54 - fts-squat: support ORs |
| 8158 | 55 - UIDVALIDITY changed while saving -> sync errors |
| 56 - mbox: copy to Trash, manually delete copied msg, change uidvalidity, | |
| 57 set nextuid=1, copy again -> error | |
| 58 - recent_uids assert at least with mbox | |
| 59 - quota fs: Should values returned by quota be divided by the actual | |
| 60 filesystem block size instead of hardcoded DEV_BSIZE? not with AIX.. | |
| 61 - mailbox list fs: Listing subscriptions with children return options doesn't | |
| 62 work unless iter_is_mailbox() returns the children flags | |
| 7502 | 63 - sieve-cmu.c crash: i_assert(buf->used - 1 == part->body_size.physical_size); |
| 7449 | 64 - convert plugin: Create a r/w lock for a file. It's read-locked if |
| 65 conversion isn't wanted and released when process dies. If conversion is | |
| 66 wanted and write-lock succeeds, conversion is done, if write-lock doesn't | |
| 67 succeed it fallbacks to using the old storage. When process is exiting it | |
| 68 again tries to write-lock and do the conversion. Add a parameter that | |
| 69 specifies if conversion should be done. | |
| 7259 | 70 - lucene: handle replacement chars? |
| 71 - squat: | |
| 7380 | 72 - wrong indexid |
| 73 - fts_build_init() assertion failed: (last_uid < last_uid_locked) | |
| 7259 | 74 - is locking done right? it reads header without file being locked? |
| 75 - split after ~8 bytes? | |
| 7391 | 76 - expunges are delayed until more mails are added |
| 7259 | 77 - test replacement chars (SEARCH / SORT / Squat) |
| 6987 | 78 |
| 7575 | 79 - dbox: |
| 80 - "File unexpectedly lost" doesn't get fixed by itself | |
| 81 - Fix support for multi-message files | |
| 7603 | 82 - Delete dovecot-keywords and dovecot-uidlist after all maildir files |
| 83 have been converted to native dbox | |
| 7449 | 84 - DEBUG: buffer overflow checking code probably doesn't handle a successful |
| 85 t_try_realloc() or pool_alloconly_realloc() properly | |
| 6987 | 86 - cache: compress when we can drop temporary fields. |
| 87 - new %modifier for reverse DNS lookups with a cache | |
| 88 - auth_gssapi_hostname = %Xl | |
| 89 - proxying would also want DNS lookups, but not reverse.. | |
| 8738 | 90 - ldap: |
| 91 - domain lookups which set the base for user lookup | |
| 92 - same attribute can't be used for multiple values. | |
| 93 - multiple attributes can't be merged to a single value. | |
| 94 - implement something like: | |
| 95 user_attrs { | |
| 96 uid = %{ldap:uidNumber} | |
| 97 home = %{ldap:homeDirectory} | |
| 98 quota_bytes = *:bytes=%{ldap:quota} | |
| 99 } | |
| 5621 | 100 |
| 7575 | 101 - Per-user options: |
| 102 - Deny deleting non-empty mailboxes | |
| 103 - Disable IDLE "still here" notifications | |
| 104 | |
| 6105 | 105 - maildir+pop3/deliver fast updates: |
| 106 - with locking enabled, pop3 could just keep the one and same sync lock and | |
| 107 do the whole thing using sync transaction | |
| 108 - don't update dovecot-uidlist if dovecot.index.cache doesn't exist / | |
| 109 there's nothing to cache | |
| 110 - if all messages are expunged and there are no unknown extensions in index, | |
| 111 unlink dovecot.index and rotate log and add some initial useful info to | |
| 112 the log (uidvalidity, nextuid) | |
| 5621 | 113 |
| 6105 | 114 - maildir |
| 6987 | 115 - don't allow more than 26 keywords |
| 6105 | 116 - physical separator could be configurable |
| 117 - maildir_copy_with_hardlinks: We're currently first hardlinking to tmp/ and | |
| 118 then rename()ing. This wouldn't be necessary if uidlist syncing noticed | |
| 119 that someone else already had added them to uidlist, and the existing UIDs | |
| 120 could be assigned to them in the index. | |
| 6754 | 121 - copying should copy already-cached data |
| 6105 | 122 - maildir_copy_preserve_filename=yes has a race condition causing "Append with |
| 123 UID n, but next_uid = y" errors when quota plugin is loaded. Practically | |
| 124 won't happen except in stress testing. | |
| 5621 | 125 |
| 6105 | 126 - mbox |
| 127 - UID renumbering doesn't really work after all? | |
| 128 - still problems with CRLF mboxes.. | |
| 129 | |
| 6987 | 130 - proxy: If remote server disconnects on login: |
| 131 login: tried to change state 2 -> 2 | |
| 6105 | 132 - logging consistency: |
| 133 http://www.dovecot.org/list/dovecot/2007-April/021532.html | |
| 134 - EXPUNGE command in read-only mailbox should give an error message if | |
| 135 there are messages marked as \Deleted? | |
| 136 - dovecot -o setting=something overriding | |
| 5557 | 137 - file_cache: we're growing the mmap in page size blocks, which is horribly |
| 138 slow if mremap() doesn't exist. | |
| 139 - login_max_processes_count shouldn't count proxying processes | |
| 140 | |
| 141 - Allow %variables in mail_chroot setting | |
| 142 - ssl_verify_client_cert isn't working if the SSL cert doesn't have CRL | |
| 143 | |
| 3087 | 144 - keywords: |
| 145 - add some limits to how many there can be | |
| 146 - don't return \* in PERMANENTFLAGS when we're full | |
| 147 - remove unused keywords? | |
| 2774 | 148 |
| 149 - caching | |
| 3904 | 150 - force bits should be used only for nonregistered fields |
| 151 - change envelope parsing not to use get_headers() so imap.envelope can | |
| 152 actually be cached without all the headers.. | |
| 6105 | 153 - if there's no other pressure for compression, we should do it when |
| 154 enough temp fields are ready to be dropped | |
| 3904 | 155 - we could try compressing same field values into a single |
| 156 location in cache file. | |
| 7259 | 157 - place some maximum limit of fields to cache file? maybe some soft and |
| 158 hard limits, so when soft limit is reached drop fields that have | |
| 159 been used only once. when hard limit is reached drop any fields to get | |
| 160 more space. all this to avoid cache file growing infinitely. | |
| 2774 | 161 |
| 3087 | 162 - mbox |
| 3904 | 163 - syncing existing indexes takes 4x longer than creating new one, why? |
| 164 - how well does dirty sync + status work? it reads the last mail every | |
| 165 time? not very good.. | |
| 166 - always add empty line. make the parser require it too? syncing should | |
| 167 make sure there always exists two LFs at end of file. raw-mbox-stream | |
| 168 should make sure the last message ends with LF even if it doesn't exist | |
| 169 in the file | |
| 170 - Quote "From ", unquote ">From " | |
| 4449 | 171 - COPY doesn't work to itself (lock assert crash, for now just disallowed) |
| 2143 | 172 |
| 173 - index | |
| 3087 | 174 - read-only support for mailboxes where we don't have write-access |
| 6105 | 175 - index file format changes: |
| 176 - pack UIDs to beginning of file with UID ranges | |
| 177 - use squat-like compressed uid ranges everywhere | |
| 6782 | 178 - write first extension intros in dovecot.index.log always with names |
| 179 - or better yet, drop the intro concept completely as it is now | |
| 180 - add "transaction boundaries" so we know which records belong to a | |
| 181 single transaction. | |
| 182 - only after that we can remove the transaction log offset | |
| 183 overwriting (otherwise we can get partial transactions in views) | |
| 2143 | 184 |
| 4449 | 185 - namespaces |
| 186 - namespaces: add new "auto_disable" flag so if the mailbox can't be opened | |
| 187 (eg. file doesn't exist), just ignore the problem and disable the | |
| 188 namespace | |
| 189 | |
| 190 - lib-storage | |
| 5557 | 191 - rename: allow moving between storages, as long as they're of same type |
| 4449 | 192 - x search charset asdf all -> should fail |
| 3087 | 193 |
| 194 - login | |
| 4449 | 195 - imap-login: Master sent reply with unknown tag 1. client closed |
| 196 connection at the exact same time master was logging it in? | |
| 197 see master_request_abort() | |
| 3087 | 198 - Digest-MD5: support integrity protection, and maybe crypting. Do it |
| 199 through login process like SSL is done? | |
| 200 - x login foo bar | |
| 201 x NO Authentication failed. | |
| 202 x login cras pass | |
| 203 * BYE Disconnected for inactivity. | |
| 204 ^ but it's not disconnecting! (buggy dovecot-auth not replying) | |
| 3744 | 205 probably because userdb lookup didn't reply, and fd was already sent |
| 206 for master.. should imap-login be handling it anymore?.. | |
| 3087 | 207 - imap-login: Authenticate PLAIN failed: Authentication failed: |
| 208 Authentication server isn't connected, try again later.. [127.0.0.1] | |
| 209 ^ NO Authentication failed. (should be Temporary login failure!) | |
| 210 - if auth process dies, login process should retry authentication if | |
| 211 possible. or if not, disconnect the client so it doesn't think the auth | |
| 212 failed. | |
| 474 | 213 |
| 3087 | 214 - auth |
| 4449 | 215 - with blocking passdb we're not caching lookups if the password was wrong |
| 5557 | 216 - non-plaintext authentication doesn't support all features: |
| 217 - multiple passdbs don't work, only the first one is used | |
| 218 - auth cache's last_success password change check doesn't exist | |
| 9121 | 219 - auth_cache_negative_ttl doesn't check password mismatches |
| 3087 | 220 - SIGHUP restarts auth processes .. but does it wait until they've finished |
| 221 with all requests? no. | |
| 222 - does dovecot-auth really break when it runs out of fds? | |
| 4449 | 223 - dovecot-auth should limit how fast authentication requests are allowed |
| 224 from login processes. especially if there's one login/connection the speed | |
| 3087 | 225 should be something like once/sec. also limit how fast to accept new |
| 226 connections. | |
| 227 - support read-only logins. user could with alternative password get only | |
| 228 read-access to mails so mails could be read relatively safely with | |
| 229 untrusted computers. Maybe always send [ALERT] about the previous | |
| 230 read-only login time with IP? | |
| 4449 | 231 - dovecot-auth workers: create a separate dovecot-pam worker which shares |
| 232 pretty much all code with dovecot-auth worker but isn't linked against | |
| 233 any libraries. or..? this might be difficult to do, especially because the | |
| 234 workers currently can handle any kind of passdb/userb requests. perhaps | |
| 235 there should be a completely separate simple PAM authenticator binary. | |
| 526 | 236 |
| 3087 | 237 - master |
| 238 - configurable syslog prefix | |
| 3450 | 239 - if there are duplicate settings, complain about it |
| 3087 | 240 |
| 241 - quota | |
| 242 - if dovecot-uidlist can't be written, assume the new mails have UIDs | |
| 243 beginning from uidlist.next_uid. Whenever mails are expunged, overwrite | |
| 244 the next_uid field with the current highest next_uid. Whenever we have | |
| 245 assumed UIDs and uidlist gets updated, throw the client out with | |
| 246 "inconsist mailbox". | |
| 247 | |
| 248 - ssl | |
| 249 - add setting: ssl_options = bitmask. by default we enable all openssl | |
| 250 workarounds, this could be used to disable some of them | |
| 251 - gnutls support isn't working | |
| 965 | 252 |
| 3087 | 253 - search |
| 6232 | 254 - message header search: we should ignore LWSP between two MIME blocks(?) |
| 255 - message_search_init() could accept multiple search keywords so we | |
| 1250 | 256 wouldn't need to call it separately for each one (so we wouldn't need |
| 257 to parse the message multiple times). | |
| 258 - could optionally support scanning inside file attachments and use | |
| 259 plugins to extract text out of them (word, excel, pdf, etc. etc.) | |
| 260 - Create our own extension: When searching with TEXT/BODY, return | |
| 261 the message text surrounding the keywords just like web search engines | |
| 262 do. like: SEARCH X-PRINT-MATCHES TEXT "hello" -> * SEARCH 1 "He said: | |
| 263 Hello world!" 2 "Hello, I'm ...". This would be especially useful with | |
| 264 the above attachment scanning. | |
| 0 | 265 |
| 3087 | 266 - general |
| 4449 | 267 - stop using atol(), atoi(), strtoul() etc. in places where we actually |
| 268 care about what they return, and rather create our own function which | |
| 269 checks if the input overflows the integer, and if so call i_fatal() | |
| 270 - LMTP server | |
| 271 - ability to build plugins statically into the binaries | |
| 3087 | 272 - ~/.dovecotrc to override system wide settings. namespace settings should |
| 273 override all the previous namespace settings instead of adding new. | |
| 274 - 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
|
275 |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
276 capabilities: |
| 3087 | 277 - 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
|
278 - mailbox-referrals (rfc2193) |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
279 - 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
|
280 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
|
281 requiring another UID to run. |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
282 - 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
|
283 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
|
284 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
|
285 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
|
286 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
|
287 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
|
288 referrals. |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
289 - 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
|
290 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
|
291 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
|
292 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
|
293 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
|
294 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
|
295 ask the password again. |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
296 - problems, problems, .. maybe not worth the trouble. |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
297 - drafts: |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
298 - http://www.imc.org/ids.html |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
299 - annotate (draft-ietf-imapext-annotate) |
|
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
300 - 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
|
301 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
|
302 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
|
303 berkeley db to store the .data file and these annotations. |
| 1178 | 304 - this is separate problem from index files. indexes are treated as |
| 305 temporary files, annotations are permanent data. we'd have to | |
| 306 support non-db way to do this too, which would probably be just a | |
| 307 simple (slow) text file. | |
| 5621 | 308 - use lib-dict probably |
| 309 - metadata (draft-daboo-imap-annotatemore) | |
|
376
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
310 - 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
|
311 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
|
312 db as well. |
| 5621 | 313 - lib-dict, again |
|
376
fd1fc4cf11b7
updated with lots of new capability comments
Timo Sirainen <tss@iki.fi>
parents:
340
diff
changeset
|
314 - binary (draft-nerenberg-imap-binary) |