1 |
Crap... I'm losing my mind... I guess I misread the "--list-increments" |
2 |
output yesterday. It appears that the backup script actually failed last |
3 |
night *and* the night before. And there's a dead.letter -- though I'd |
4 |
swear it wasn't there yesterday -- that says the script still couldn't |
5 |
find my exclude-list.txt. I changed the script to give it the full path |
6 |
to that file... hopefully that'll fix it. |
7 |
|
8 |
Sorry for the wild goose chase... |
9 |
|
10 |
b |
11 |
|
12 |
|
13 |
Richard Yumul wrote: |
14 |
> are you specifying the full path to logger? The environment when a |
15 |
> script is run from cron is not always the same as the root env., so you |
16 |
> typically have to specify the full path to the program... |
17 |
> |
18 |
> HTH, |
19 |
> Rich |
20 |
> |
21 |
> |
22 |
> |
23 |
> Ben Munat wrote: |
24 |
> |
25 |
>> What's interesting is that cron ran my backup last night |
26 |
>> successfully... but didn't send the output to the log. |
27 |
>> |
28 |
>> I have lines like "logger "Server backup: blah blah blah"" in my |
29 |
>> script and I pipe the rdiff-backup output into logger as well. When I |
30 |
>> ran the script from the command line, the output showed up just fine |
31 |
>> in the log. But, when cron runs it, not a peep... and no dead.letter |
32 |
>> either. |
33 |
>> |
34 |
>> Oh well, at least my nightly backup is automated now. |
35 |
>> |
36 |
>> b |
37 |
>> |
38 |
>> |
39 |
>> Jacob Joseph wrote: |
40 |
>> |
41 |
>>> I was watching this conversation to see how others do this sort of |
42 |
>>> thing and am glad to have learned of dead.letter. Thanks. |
43 |
>>> |
44 |
>>> Incidentally, I typically do this sort of debugging by just |
45 |
>>> redirecting output to a file. That is: |
46 |
>>> |
47 |
>>> * * * * * root <script> > /tmp/foo 2>&1 |
48 |
>>> |
49 |
>>> It always seems like the easiest way to go. |
50 |
>>> |
51 |
>>> -Jacob |
52 |
>>> |
53 |
>>> Aaron Walker wrote: |
54 |
>>> |
55 |
>>>> -----BEGIN PGP SIGNED MESSAGE----- |
56 |
>>>> Hash: SHA1 |
57 |
>>>> |
58 |
>>>> Ben Munat wrote: |
59 |
>>>> |
60 |
>>>> | PS: or maybe there's somewhere that cron puts the output when it |
61 |
>>>> can't |
62 |
>>>> | mail it? |
63 |
>>>> |
64 |
>>>> If you're running the script as root, mail it couldn't deliver would |
65 |
>>>> be put in |
66 |
>>>> /root/dead.letter. |
67 |
>>>> |
68 |
>>>> | |
69 |
>>>> | Jonathan Rogers wrote: |
70 |
>>>> | |
71 |
>>>> |> |
72 |
>>>> |> Yes, the programs in /etc/cron.* are running as root, but |
73 |
>>>> probably not |
74 |
>>>> |> with exactly the same environment as when you run them from a login |
75 |
>>>> |> shell. The cron daemon should mail the output from your script if it |
76 |
>>>> |> fails. If you can get the error messages, I might be able to help. |
77 |
>>>> |> |
78 |
>>>> |> Jonathan Rogers |
79 |
>>>> |> |
80 |
>>>> |
81 |
>>>> |
82 |
>>>> - -- |
83 |
>>>> Eat drink and be merry, for tomorrow we diet. |
84 |
>>>> |
85 |
>>>> Aaron Walker < ka0ttic@g.o > http://dev.gentoo.org/~ka0ttic/ |
86 |
>>>> Gentoo/BSD | cron | shell-tools |
87 |
>>>> http://butsugenjitemple.org/~ka0ttic/ |
88 |
>>>> |
89 |
>>>> -----BEGIN PGP SIGNATURE----- |
90 |
>>>> Version: GnuPG v1.2.6 (GNU/Linux) |
91 |
>>>> Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org |
92 |
>>>> |
93 |
>>>> iD8DBQFBlloKC3poscuANHARAja4AJ9qDNISjFWZlpDFd4jIzobQ8uh3rQCguO1t |
94 |
>>>> y5+lQ06YyqeK58HvD2zRCpY= |
95 |
>>>> =YsZz |
96 |
>>>> -----END PGP SIGNATURE----- |
97 |
>>> |
98 |
>>> |
99 |
>>> |
100 |
> |
101 |
> -- |
102 |
> ------------------------------------------------------------------------ |
103 |
> Richard M Yumul |
104 |
> rmy@×××××××××.com <mailto:rmy@×××××××××.com> |
105 |
> SDTechnix |
106 |
> http://www.sdtechnix.com |
107 |
> ------------------------------------------------------------------------ |