1 |
On Mon, 15 Mar 2010 23:20:03 +0100, Willie Wong wrote about Re: |
2 |
[gentoo-user] what's wrong with rsync 3.0.6?: |
3 |
|
4 |
>On Mon, Mar 15, 2010 at 05:08:06PM +0000, David W Noon wrote: |
5 |
[snip] |
6 |
>> Because emerge jobs produce copious amounts of output that is |
7 |
>> difficult to read as it scrolls past. I much prefer the cron daemon |
8 |
>> or at daemon to send me the output as email, so I can scroll |
9 |
>> backwards and forwards through it at my leisure. |
10 |
> |
11 |
>What output do you actually read from syncs? |
12 |
|
13 |
I look at the differences in the Portage tree before and after. |
14 |
|
15 |
>For builds, is it really wise to do all updates unattended? |
16 |
|
17 |
Perfectly. The emerge runs the same whether in background or |
18 |
foreground. If it's going to trash something, it will do it the same |
19 |
way whether you use an "at" job or run it directly in the console. |
20 |
|
21 |
> Also, for builds, there is such a |
22 |
>thing as elogs (which allows you to save all messages to |
23 |
>/var/log/portage for ease of reading at your leisure. |
24 |
|
25 |
I have mine go to /var/log/portage/log. But these only log the |
26 |
activities within a single ebuild, not the other housekeeping that goes |
27 |
on in an emerge job. The output of a batch job contains the lot. |
28 |
|
29 |
>I'm sure you have a good reason for wanting to do things your way, and |
30 |
>I do not claim mine is better. I am just surprised that you sounded |
31 |
>surprised to find out some people don't do things your way. |
32 |
|
33 |
Perhaps it is because I became used to running long-winded jobs in the |
34 |
background years ago on mainframes. It was always the case that using |
35 |
the terminal as the primary output device slowed down the job, because |
36 |
spooling the output to disk was always faster than displaying it on the |
37 |
terminal. I believe that to be so to this very day. |
38 |
|
39 |
>Actually, the cron daemon won't run because I don't have a cron daemon |
40 |
>installed on the laptop. And I don't have a cron daemon because having |
41 |
>periodic jobs only make sense if the computer is likely to be on when |
42 |
>cron is triggered. |
43 |
|
44 |
Using a modern cron daemon is a convenient way to run regular jobs, |
45 |
even on a machine that is powered off for much of the time. One can |
46 |
use the "first" option to kick off jobs relative to power-on time |
47 |
instead of wall clock time. |
48 |
-- |
49 |
Regards, |
50 |
|
51 |
Dave [RLU #314465] |
52 |
====================================================================== |
53 |
dwnoon@××××××××.com (David W Noon) |
54 |
====================================================================== |