Mercurial > dovecot > core-2.2
annotate TODO @ 12755:baa0314ed610
Merged changes from v2.0 tree.
| author | Timo Sirainen <tss@iki.fi> |
|---|---|
| date | Tue, 22 Feb 2011 15:31:31 +0200 |
| parents | a2780b694b2d 440fcf8cb338 |
| children | 04073884748a |
| rev | line source |
|---|---|
|
12586
a2780b694b2d
lib-storage: mailbox_alloc() now takes a virtual mailbox name and other related API changes.
Timo Sirainen <tss@iki.fi>
parents:
12579
diff
changeset
|
1 - check: |
|
a2780b694b2d
lib-storage: mailbox_alloc() now takes a virtual mailbox name and other related API changes.
Timo Sirainen <tss@iki.fi>
parents:
12579
diff
changeset
|
2 - dsyncing between two namespace separators is probably broken.. |
| 12579 | 3 - remove mail_deliver_session after all, do all the stuff transparently |
| 4 by hooking into mailbox_copy(). | |
| 5 - use this hook also to do the mail deduplication: 1) sort all destination | |
| 6 users, 2) create mail_user only once for each user, 3) remember in | |
| 7 src_mail the previously copied mail, 4) use that for mailbox_copy()ing | |
| 8 to following recipients | |
| 12551 | 9 - notify_sync() could have "what changed" struct with old/new flags |
| 12546 | 10 - mailbox deletion could really be "rmdir", which fails if everything isn't |
| 11 already expunged. IMAP DELETE could fail with "mailbox changed during | |
| 12 deletion, aborting" if new messages arrive | |
| 12551 | 13 - if doing any changes, figure out how to handle symlinked mailboxes |
| 12546 | 14 - maildir: copy dovecot-shared file from parent mailbox, not root. |
| 15 - is mail_index_sync_have_any() really necessary? especially that it returns | |
| 16 TRUE for ext rec changes ... why?... | |
| 17 - mailbox listing to use UTF8 everywhere | |
| 18 - move namespace sep vs. real_sep check from imap client_find_namespace() | |
| 19 to lib-storage mailbox name validity check | |
| 20 - rewrite autocreate plugin by making it list virtual mailbox names and only | |
| 21 lazily autocreate create them when mailbox is opened | |
| 22 - login-common API change to fix compiling with | |
| 23 LDFLAGS="-Wl,--as-needed -Wl,--no-undefined" | |
| 24 - if global ACL path points to a file instead of a directory, read all of | |
| 25 them from there. support wildcards in names. | |
| 26 - remove 3 unnecessary stat() calls when opening mailbox. they all come from | |
| 27 mailbox_list_get_permissions_full() | |
| 28 | |
| 12476 | 29 - master passdb preserves userdb_* extra fields. should it preserve |
| 30 non-userdb_* extra fields too? | |
| 12400 | 31 - imap, pop3: if client init fails, wait a second or two before disconnecting |
| 32 client. | |
| 33 - doveadm expunge parameter to delete empty mailboxes (for lazy-expunge) | |
| 34 - doveadm search savedbefore 7d could be optimized in large mailboxes.. | |
| 12336 | 35 - mdbox: storage rebuilding could log about changes it does |
| 12400 | 36 - mdbox: broken extrefs header keeps causing index rebuilds |
| 12336 | 37 - cache: mail_cache_lookup() should be able to return mail_cache_add()ed |
| 38 fields even before they've been flushed to disk. this is useful when copying | |
| 39 messages and some plugin (e.g. mail_log) fetches some fields that are | |
| 40 already added to cache (to avoid opening and parsing the message) | |
| 12231 | 41 - doveadm -A <<EOF expunge + purge + EOF |
| 42 - sent, drafts: .Sent/dovecot.index: modseq_hdr.log_offset too large | |
| 12181 | 43 - mail_max_lock_timeout error could be reported more nicely, also ones coming |
| 44 from lib-index | |
| 45 - dsync: if mailbox is renamed due to it being invalid, its subscription | |
| 46 isn't renamed | |
| 12132 | 47 - doveadm director assign <user> <host> - do it permanently in memory |
| 48 - LSUB shows subscribed namespace prefix as prefix/, also SUBSCRIBE adds the / | |
| 12095 | 49 - sql pool: if async query is pending and sync query is sent and there |
| 50 are no more empty connections, it should flush the async query first | |
| 11683 | 51 - NTLMv1 and LM should be disabled if disable_plaintext_auth=yes |
| 11387 | 52 - SEARCH SENT*/HEADER/etc. doesn't seem optimized when using with TEXT/BODY |
| 12476 | 53 - dict sql: support ignoring some search key hierarchies (e.g. acl "anyone") |
| 11273 | 54 |
| 11683 | 55 - fts-solr: crashes if expunge is done while search is indexing |
| 10968 | 56 - fts-solr: handle DELETE, RENAME. use mailbox GUIDs (optionally) |
| 10546 | 57 |
| 10479 | 58 - mdbox |
| 11485 | 59 - dotlocking: cleanup should delete stale *.lock files |
| 10968 | 60 - purging seems to be inefficient. run imaptest for a while, get >500 |
| 61 files, start purging, it's slow until there are about 100 files left, | |
| 62 then the rest is suddenly fast. | |
| 63 - make sure that when reading mdbox mails sequentially the data is being | |
| 64 read from disk in n kB blocks and reads cross mail boundaries and when | |
| 65 reading the next mail it uses the previously read data in buffer | |
| 10479 | 66 - Add some kind of checksum about data+metadata and use it when checking |
| 67 consistency | |
| 10798 | 68 - figure out a way to efficiently trigger purging when user has too much |
| 69 mail expunged (e.g. keep track of total storage size, trigger purging | |
| 70 when it's 2*quota limit) | |
| 71 - keep track of total bytes in dbox storage in map header. also if | |
| 72 possible keep track of refcount=0 bytes. use these to optimize checks. | |
| 73 - save some stuff to map index header so we don't need to keep retrying | |
| 74 it. like when saving the lowest file_id which to bother checking. | |
| 75 - test crash-fixing | |
| 76 - mail_index_fsck() should perhaps cause dbox to be resynced? | |
| 10968 | 77 - optimize away reading file header? |
| 10479 | 78 - maildir: out-of-disk-space failures apparently cause all kinds of |
| 79 problems, e.g. "Expunged message reappeared", "Duplicate file entry"? | |
| 10163 | 80 - deliver -r <address> used as autoreplies' From-address? |
| 81 - add fd limit checks/warnings | |
| 10546 | 82 - istream-seekable is inefficient. it shouldn't be reading the temp file |
| 83 immediately after writing to it | |
| 10968 | 84 - config process is handling requests too slowly. maybe add some caching. |
| 85 - maybe config should return all of the protocol/local/remote overrides | |
| 86 when requested? then the caller could do a single lookup at start and | |
| 87 merge them later internally. this would really help login processes. | |
| 88 - ipv6: auth penalty should begin from /64 and gradually grow to /48 if | |
| 89 necessary. and the same could be done for ipv4 as well.. | |
| 10052 | 90 |
| 9559 | 91 - dsync: |
| 11683 | 92 - check for uid conflicts in the middle of mailbox based on next existing |
| 93 message, not previous | |
| 10479 | 94 - subscriptions syncing doesn't handle mixed hierarchy separators |
| 10798 | 95 - Can't rename mailbox INBOX_019acf2169c5784b307f000074ccac23 to INBOX: Target mailbox already exists |
| 11683 | 96 - copying can't work in non-full mode unless source mailbox is modified |
| 97 in some way. possibly add some COPY records to transaction log in future? | |
| 98 - superfast mode: when syncing changes within a mailbox, use modseqs to | |
| 99 figure out what has changed similar to QRESYNC. | |
| 10052 | 100 |
| 10479 | 101 - lda: add some -h 'Header: value' parameter that adds/replaces header |
| 10163 | 102 - ldap: fix multiple-gid support somehow |
| 9459 | 103 - search: use mail_get_parts() only when it's already cached. if it's not, |
| 104 add it to cache afterwards. | |
|
9235
2e2b957f1cca
Implemented anvil service, which is used to implement mail_max_userip_connections.
Timo Sirainen <tss@iki.fi>
parents:
9159
diff
changeset
|
105 - move ssl proxying code to lib-master |
| 10798 | 106 - proxy: verify ssl cert's cname |
|
9004
a5555606846b
Fixed plugins to work with config rewrite.
Timo Sirainen <tss@iki.fi>
parents:
9002
diff
changeset
|
107 - dict pooling |
| 9015 | 108 |
| 9000 | 109 /* currently non-external transactions can be applied multiple times, |
| 110 causing multiple increments. */ | |
| 111 //FIXME:i_assert((t->flags & MAIL_INDEX_TRANSACTION_FLAG_EXTERNAL) != 0); | |
| 112 ^ appears to work now though, probably because of the added syncing stuff.. | |
| 113 | |
| 114 - transaction log corruption should make sure dovecot.index is rewritten | |
| 115 and perhaps not delete the file. | |
| 116 - use backup index in mail_index_fsck() | |
| 117 | |
|
8564
3c4934783aff
Mention SSL/TLS in "plaintext auth disallowed" error.
Timo Sirainen <tss@iki.fi>
parents:
8555
diff
changeset
|
118 - 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
|
119 one is down |
| 8555 | 120 - i_panic("Message count decreased") happens - why? |
|
8714
bbdbab5354d6
Implemented support for ESORT capability.
Timo Sirainen <tss@iki.fi>
parents:
8564
diff
changeset
|
121 - at least one backtrace shows client_destroy -> client_command_cancel -> |
|
bbdbab5354d6
Implemented support for ESORT capability.
Timo Sirainen <tss@iki.fi>
parents:
8564
diff
changeset
|
122 imap_sync_deinit |
| 8555 | 123 - fsck -> log_file_tail_offset 2273345664 -> 996 -> |
| 124 mail-transaction-log.c: line 341 (mail_transaction_log_set_mailbox_sync_pos): | |
| 125 assertion failed: (file_offset >= log->head->saved_tail_offset) | |
| 11765 | 126 - virtual: If last message matching INTHREAD rule gets expunged, the rest of |
| 127 the thread doesn't go away | |
| 8488 | 128 - virtual: "Searched n% of the mailbox" gives broken numbers since |
| 8555 | 129 ctx->seq jumps around. And why is it also returned when fts is enabled |
| 130 along with "Indexed n% of the mailbox"? | |
| 8488 | 131 - how do shared mailboxes work with plugins? |
| 132 - lazy-expunge, fts, etc.? | |
| 11974 | 133 - listescape+acl can't handle shared mailboxes with escape chars |
| 8488 | 134 - dovecot-acl-list: |
| 135 - how does it work with global acls? | |
| 136 - update immediately after SETACL: add/remove entries, update timestamps | |
| 137 - read the entire file to memory only once and keep it there, stat() later | |
| 138 to see if it has changed. if not, perhaps don't even bother stat()ing | |
| 139 dovecot-acl files? at least not that often.. | |
| 140 - fs quota: getquotaroot inbox vs. other-box should return different quotas | |
| 141 if two quotas are defined | |
| 8158 | 142 - new primes code: are hash tables now being resized too often? |
| 143 - auth_log_prefix setting similar to mail_log_prefix | |
| 144 | |
| 145 - thread indexes: if we expunge a duplicate message-id: and we have a sibling | |
| 146 with identical message-id:, we can probably just move the children? | |
| 147 (unless there are non-sibling duplicates) | |
| 148 - SEARCH INTHREAD requires no thread sorting by date - don't do it | |
| 149 - CONDSTORE: use per-flag/per-keyword conflict checking | |
| 150 - QRESYNC: Drop expunges from the middle of given seq sets if possible | |
| 151 - use universal hash functions? | |
| 152 | |
| 153 - UIDVALIDITY changed while saving -> sync errors | |
| 154 - mbox: copy to Trash, manually delete copied msg, change uidvalidity, | |
| 155 set nextuid=1, copy again -> error | |
| 156 - recent_uids assert at least with mbox | |
| 157 - quota fs: Should values returned by quota be divided by the actual | |
| 158 filesystem block size instead of hardcoded DEV_BSIZE? not with AIX.. | |
| 159 - mailbox list fs: Listing subscriptions with children return options doesn't | |
| 160 work unless iter_is_mailbox() returns the children flags | |
| 7259 | 161 - lucene: handle replacement chars? |
| 162 - squat: | |
| 11273 | 163 - support ORs |
| 7380 | 164 - wrong indexid |
| 165 - fts_build_init() assertion failed: (last_uid < last_uid_locked) | |
| 7259 | 166 - is locking done right? it reads header without file being locked? |
| 167 - split after ~8 bytes? | |
| 7391 | 168 - expunges are delayed until more mails are added |
| 7259 | 169 - test replacement chars (SEARCH / SORT / Squat) |
| 6987 | 170 |
| 7449 | 171 - DEBUG: buffer overflow checking code probably doesn't handle a successful |
| 172 t_try_realloc() or pool_alloconly_realloc() properly | |
| 8738 | 173 - ldap: |
| 12400 | 174 - support multiple connections for doing auth binds |
| 8738 | 175 - domain lookups which set the base for user lookup |
| 176 - same attribute can't be used for multiple values. | |
| 177 - multiple attributes can't be merged to a single value. | |
| 12400 | 178 - multiple value could be joined with specified separator (per-field) |
| 8738 | 179 - implement something like: |
| 180 user_attrs { | |
| 181 uid = %{ldap:uidNumber} | |
| 182 home = %{ldap:homeDirectory} | |
| 183 quota_bytes = *:bytes=%{ldap:quota} | |
| 184 } | |
| 5621 | 185 |
|
9159
6324a79d3ee1
Initial commit for v2.0 master rewrite. Several features are still missing.
Timo Sirainen <tss@iki.fi>
parents:
9015
diff
changeset
|
186 - maildir+pop3 fast updates: |
| 6105 | 187 - with locking enabled, pop3 could just keep the one and same sync lock and |
| 188 do the whole thing using sync transaction | |
| 189 - don't update dovecot-uidlist if dovecot.index.cache doesn't exist / | |
| 190 there's nothing to cache | |
| 191 - if all messages are expunged and there are no unknown extensions in index, | |
| 192 unlink dovecot.index and rotate log and add some initial useful info to | |
| 193 the log (uidvalidity, nextuid) | |
| 5621 | 194 |
| 6105 | 195 - maildir |
| 6987 | 196 - don't allow more than 26 keywords |
| 6105 | 197 - physical separator could be configurable |
|
9159
6324a79d3ee1
Initial commit for v2.0 master rewrite. Several features are still missing.
Timo Sirainen <tss@iki.fi>
parents:
9015
diff
changeset
|
198 - lda+maildir: if new mails are in new/ or cur/ they're not added to |
| 6105 | 199 dovecot-uidlist but newly saved mails are, so UIDs will be in wrong order |
| 5621 | 200 |
| 5557 | 201 - file_cache: we're growing the mmap in page size blocks, which is horribly |
| 202 slow if mremap() doesn't exist. | |
| 203 | |
| 3087 | 204 - keywords: |
| 205 - add some limits to how many there can be | |
| 206 - don't return \* in PERMANENTFLAGS when we're full | |
| 207 - remove unused keywords? | |
| 2774 | 208 |
| 11273 | 209 - mail caching |
| 3904 | 210 - force bits should be used only for nonregistered fields |
| 211 - change envelope parsing not to use get_headers() so imap.envelope can | |
| 212 actually be cached without all the headers.. | |
| 6105 | 213 - if there's no other pressure for compression, we should do it when |
| 214 enough temp fields are ready to be dropped | |
| 3904 | 215 - we could try compressing same field values into a single |
| 216 location in cache file. | |
| 7259 | 217 - place some maximum limit of fields to cache file? maybe some soft and |
| 218 hard limits, so when soft limit is reached drop fields that have | |
| 219 been used only once. when hard limit is reached drop any fields to get | |
| 220 more space. all this to avoid cache file growing infinitely. | |
| 2774 | 221 |
| 3087 | 222 - mbox |
| 11273 | 223 - UID renumbering doesn't really work after all? |
| 12476 | 224 - still problems with CRLF mboxes.. especially with broken Content-Length |
| 225 headers (pointing between CR-LF?) | |
| 3904 | 226 - syncing existing indexes takes 4x longer than creating new one, why? |
| 227 - how well does dirty sync + status work? it reads the last mail every | |
| 228 time? not very good.. | |
| 229 - always add empty line. make the parser require it too? syncing should | |
| 230 make sure there always exists two LFs at end of file. raw-mbox-stream | |
| 231 should make sure the last message ends with LF even if it doesn't exist | |
| 232 in the file | |
| 233 - Quote "From ", unquote ">From " | |
| 4449 | 234 - COPY doesn't work to itself (lock assert crash, for now just disallowed) |
| 2143 | 235 |
| 236 - index | |
| 3087 | 237 - read-only support for mailboxes where we don't have write-access |
| 6105 | 238 - index file format changes: |
| 10968 | 239 - split to "old" and "new" indexes and try to avoid loading "old" into |
| 240 memory until needed | |
| 6105 | 241 - pack UIDs to beginning of file with UID ranges |
| 242 - use squat-like compressed uid ranges everywhere | |
| 6782 | 243 - write first extension intros in dovecot.index.log always with names |
| 244 - or better yet, drop the intro concept completely as it is now | |
| 245 - add "transaction boundaries" so we know which records belong to a | |
| 246 single transaction. | |
| 247 - only after that we can remove the transaction log offset | |
| 248 overwriting (otherwise we can get partial transactions in views) | |
| 2143 | 249 |
| 4449 | 250 - lib-storage |
| 251 - x search charset asdf all -> should fail | |
| 3087 | 252 |
| 253 - login | |
| 254 - Digest-MD5: support integrity protection, and maybe crypting. Do it | |
| 255 through login process like SSL is done? | |
| 474 | 256 |
| 3087 | 257 - auth |
| 4449 | 258 - with blocking passdb we're not caching lookups if the password was wrong |
| 5557 | 259 - non-plaintext authentication doesn't support all features: |
| 260 - multiple passdbs don't work, only the first one is used | |
| 261 - auth cache's last_success password change check doesn't exist | |
| 9459 | 262 - auth_cache_negative_ttl doesn't check password mismatches |
| 4449 | 263 - dovecot-auth should limit how fast authentication requests are allowed |
| 264 from login processes. especially if there's one login/connection the speed | |
| 3087 | 265 should be something like once/sec. also limit how fast to accept new |
| 266 connections. | |
| 267 - support read-only logins. user could with alternative password get only | |
| 268 read-access to mails so mails could be read relatively safely with | |
| 269 untrusted computers. Maybe always send [ALERT] about the previous | |
| 270 read-only login time with IP? | |
| 271 | |
| 272 - quota | |
| 273 - if dovecot-uidlist can't be written, assume the new mails have UIDs | |
| 274 beginning from uidlist.next_uid. Whenever mails are expunged, overwrite | |
| 275 the next_uid field with the current highest next_uid. Whenever we have | |
| 276 assumed UIDs and uidlist gets updated, throw the client out with | |
| 277 "inconsist mailbox". | |
| 278 | |
| 279 - ssl | |
| 280 - add setting: ssl_options = bitmask. by default we enable all openssl | |
| 281 workarounds, this could be used to disable some of them | |
| 282 - gnutls support isn't working | |
| 965 | 283 |
| 3087 | 284 - search |
| 6232 | 285 - message header search: we should ignore LWSP between two MIME blocks(?) |
| 286 - message_search_init() could accept multiple search keywords so we | |
| 1250 | 287 wouldn't need to call it separately for each one (so we wouldn't need |
| 288 to parse the message multiple times). | |
| 289 - could optionally support scanning inside file attachments and use | |
| 290 plugins to extract text out of them (word, excel, pdf, etc. etc.) | |
| 291 - Create our own extension: When searching with TEXT/BODY, return | |
| 292 the message text surrounding the keywords just like web search engines | |
| 293 do. like: SEARCH X-PRINT-MATCHES TEXT "hello" -> * SEARCH 1 "He said: | |
| 294 Hello world!" 2 "Hello, I'm ...". This would be especially useful with | |
| 295 the above attachment scanning. | |
| 0 | 296 |
| 3087 | 297 - general |
| 298 - things break if next_uid gets to 2^32 |