File: FAQ

package info (click to toggle)
lurker 2.1-7
  • links: PTS
  • area: main
  • in suites: etch, etch-m68k
  • size: 1,800 kB
  • ctags: 1,088
  • sloc: cpp: 10,527; sh: 1,348; xml: 1,278; perl: 206; makefile: 198
file content (170 lines) | stat: -rw-r--r-- 7,043 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
Why are the dates broken on messages imported from a maildir?
Why is lurker warning about From lines during my import?
Why are new mails not appearing in my archive?
How do I add support for my language to lurker?
Why are all my dates off by a fixed hour / don't use "From " header?
Why shouldn't I open the lurker mboxs with a mail reader?
Why do I get "opening database: Operation not supported" on MacOS?
I run postfix and lurker doesn't import any new email; why?
Why not store all the messages as XML instead of rendering them as it?
How can I delete a mailing list?

---

Q: Why are the dates broken on messages imported from a maildir?

A: You probably imported from a maildir which was improperly copied/created.
   
   Normal 'cp' will reset the modification time of messages. With the
   maildir format, the modification time in new/ is the delivery time. 
   When copying a maildir use 'tar' or 'cp -a' to properly preserve this
   information.
   
   Another problem are mailbox to maildir conversion scripts. Most seem to
   have been written by people with an incomplete understanding of the
   maildir and/or mailbox formats. You should never use lurker in
   combination with a conversion scripts. Lurker has very good mailbox
   support and should import directly from the original copy.
   

Q: Why is lurker warning about From lines during my import?

A: Lurker tries to detect corrupt mailboxes. If the quoted text included
   in the warning was not the start of a new message, then lurker just
   fixed your mailbox for you.
   
   On the other hand, if the quoted text was a message boundary, you have
   a badly corrupted mailbox. You will have to repair this mailbox somehow.
   What lurker choose to do in this case was combine those two messages into
   a single message which was not correct. You should fix the mailbox, purge
   your database, and reimport. Do not import corrupted mailboxes!
   
   Be aware that when importing with '-m' lurker will always treat lines
   starting with 'From ' as part of the message body without a warning.


Q: Why are new mails not appearing in my archive?

A: You probably forgot to setup a cronjob to run lurker-prune.
   Please read the INSTALL file, section 6.


Q: How do I add support for my language to lurker?

A: First, find out your ISO 639 language code.
   All the valid ISO 639 language codes are contained in lang.xml.
   If your language does not have a standardized country code, contact
   lurker-users and express your intent to translate and ask what to do.
   
   Check the entry in lang.xml and confirm that your language's name is
   spelt correctly in the language you are translating to. Change
   localized="no" to localized="yes".
   
   Copy en.xml to xx.xml were xx is the country code of the new language.
   Edit xx.xml replacing the quoted ("") english fields with a UTF-8
   encoding of the same string, but in the new language.
   
   Check that the translation works by loading the lurker front-page and
   exploring all the possible types of lurker web-pages.
   
   Send your freshly created file to lurker-users@lists.sourceforge.net
   so that you will be included in the lurker credits and other people
   can benefit from your work!


Q: Why are all my dates off by a fixed hour / don't use "From " header?

A: The computer which imports mail should be in the same timezone as the
   computer which archived the mail. This is because mailbox timestamp
   envelopes are generated in local time. You can set your local timezone
   for the purposes of indexing with the TZ environment variable.


Q: Why shouldn't I open the lurker mboxs with a mail reader?

A: Mail readers like to modify the X-Status header by adding or removing
   flags which denote that the message is new, or opened, or read, etc.
   These changes result in displacement changes in the file even if you do
   nothing but read the file. Also, your mail reader may re-order messages
   by some criterea. Finally, you might accidentally delete a message.
   
   Solutions:
     1. Use a procmail rule to copy mail to a lurker mbox and your own mbox
        which you delete messages from:
          :0 c
          lurkers-copy.mbox
     2. Make the mailbox read only for you, but appendable by the mail
        user (or whatever user appends to the mbox):
          chown mail.youruser lurkers.mbox
          chmod 0640 lurkers.mbox


Q: Why do I get "opening database: Operation not supported" on MacOS?

A: You are running the lurker database on a filesystem which does not
   support file locking. Lurker cannot run safely on such a filesystem and
   thus refuses to possibly corrupt the database.


Q: I run postfix and lurker doesn't import any new email; why?

A: Postfix sets a file size resource limit. The intention of this option
   is to limit the size of mailboxes delivered to. Unfortunately, this
   option also applies to ALL files opened by child processes.
   
   This means lurker will die with SIGXFSZ when accessing the database!
   You must tell postfix not to set a file limit or use another MTA.
   
   mailbox_size_limit = 0
   
   If you have a postfix with a soft limit (does not exist yet), then
   compile the stand-alone file lurker-drop-rlimit.cpp and run it infront of
   lurker-index like:
   
      lurker-drop-rlimit lurker-index -c /etc/lurker.conf -l devel -m

 
Q: Why not store all the messages as XML instead of rendering them?

A: Lurker is designed to hold lots of email. Many other programs need
   mailing list archives in mbox format. It doesn't cost much for me to
   re-parse a specific email, so there is not much speed lost. On the 
   other hand, I save a lot of disk space.
   
   Example of archivers that use mbox->html + indexer:
	1Gb mbox
	2Gb html (larger on average)
	500Mb index
	---
	250% space wasted

   Example of lurker:
	1Gb mbox
	300Mb index
	---
	30% space wasted

   As you can see, since lurker builds the index from the *mbox* the index
   should be smaller. Furthermore, why store the larger html/xml when most
   of it probably won't be accessed?
   
   Finally, even though I like XML, I think it is important to be able to
   retreive to original email without any concerns that translation back and
   forth may have slightly altered the contents of the email. Therefore, I
   consider keeping the mbox essential, and it seems natural to leave the
   data in this native format.


Q: How can I delete a mailing list?

A: Bad  answer: 'lurker-search -c lurker.conf -d ml:listid -v'
   Good answer: delete the mailbox and reimport.
   
   You can certainly perform the delete command shown above, but
   cross-posted emails will then also be deleted. Furthermore, if you are
   deleting that many messages, you probably want your disk space back.
   
   A better solution is to remove the mailing list entry in the lurker.conf,
   delete it's mailbox in the database directory, and run lurker-regenerate.
   Cross-posted messages will remain in the database, and you will recover
   your disk space.