1 |
On Friday, June 06, 2014 04:46:35 AM covici@××××××××××.com wrote: |
2 |
> J. Roeleveld <joost@××××××××.org> wrote: |
3 |
> > On Friday, June 06, 2014 03:45:17 AM covici@××××××××××.com wrote: |
4 |
> > > J. Roeleveld <joost@××××××××.org> wrote: |
5 |
> > > > On Friday, June 06, 2014 01:59:18 AM covici@××××××××××.com wrote: |
6 |
> > > > > Hi. I am having some strange performance problems when booted under |
7 |
> > > > > systemd. These problems happened a little bit under openrc, but are |
8 |
> > > > > much more pronounced with systemd. |
9 |
> > > > |
10 |
> > > > I don't think it's necessarily systemd itself, just a setting that |
11 |
> > > > systemd |
12 |
> > > > does differently then openrc. See below for more. |
13 |
> > > > |
14 |
> > > > > I am using just virtual consoles, no gui whatsoever at the moment. |
15 |
> > > > > I |
16 |
> > > > > also use tmux with 4 windows in one of the vcs. My system is an i7 |
17 |
> > > > > processor, quod core and 16g of ram and 2g of swap space which |
18 |
> > > > > appears |
19 |
> > > > > not to be used. I am using uvesafb for the console, so I get 64x160 |
20 |
> > > > > screens. |
21 |
> > > > |
22 |
> > > > Sounds similar to my laptop, except I run KDE and got 16g of swap (for |
23 |
> > > > hibernate) |
24 |
> > > > |
25 |
> > > > > The first problem is that if I don't press any keystrokes for |
26 |
> > > > > several |
27 |
> > > > > minutes and then want to move to another vc, it takes about 3 or 4 |
28 |
> > > > > seconds after the alt-left arrow or alt-right arrow command to take |
29 |
> > > > > effect. Even within the same vt, if I don't do anything for several |
30 |
> > > > > minutes, it takes several seconds till the keystroke echoes and |
31 |
> > > > > something happens. Once I have done this, things act normally, but |
32 |
> > > > > its |
33 |
> > > > > kind of annoying. |
34 |
> > > > |
35 |
> > > > Sounds like a powersave setting. I used to get the same on my old |
36 |
> > > > laptop |
37 |
> > > > with spinning rust. SSDs tend to "spin-up" a lot quicker. |
38 |
> > > > |
39 |
> > > > > Also, my load average seems to always be >1. I have |
40 |
> > > > > looked at top and things seem to be OK, except that my cpu usage is |
41 |
> > > > > like |
42 |
> > > > > this: |
43 |
> > > > > Tasks: 934 total, 2 running, 931 sleeping, 0 stopped, 1 zombie |
44 |
> > > > > %Cpu(s): 12.5 us, 1.2 sy, 0.0 ni, 86.0 id, 0.2 wa, 0.0 hi, 0.0 |
45 |
> > > > > si, |
46 |
> > > > > 0.0 st |
47 |
> > > > > KiB Mem: 16450248 total, 9678656 used, 6771592 free, 1084088 |
48 |
> > > > > buffers |
49 |
> > > > > KiB Swap: 2097148 total, 4 used, 2097144 free. 1147688 |
50 |
> > > > > cached |
51 |
> > > > > Mem |
52 |
> > > > > |
53 |
> > > > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ |
54 |
> > > > > COMMAND |
55 |
> > > > > |
56 |
> > > > > 9969 root 20 0 708 16 0 R 100.0 0.0 1549:10 |
57 |
> > > > > v86d |
58 |
> > > > > |
59 |
> > > > > 579 root 30 10 0 0 0 S 9.1 0.0 16:09.93 |
60 |
> > > > > speakup |
61 |
> > > > > |
62 |
> > > > > 11789 root 20 0 22524 2388 1116 R 0.7 0.0 0:00.03 |
63 |
> > > > > top |
64 |
> > > > > |
65 |
> > > > > 7 root 0 -20 0 0 0 S 0.3 0.0 0:10.41 |
66 |
> > > > > |
67 |
> > > > > kworker/u:0H |
68 |
> > > > > |
69 |
> > > > > and onward ... |
70 |
> > > > > This is an awful lot of tasks, I have never seen so many! |
71 |
> > > > |
72 |
> > > > That is a lot, I am currently running KDE, firefox and a citrix remote |
73 |
> > > > desktop thing. (oh, and skype and kopete and a few other items) |
74 |
> > > > KDE is installed with semantic-desktop, but the nepomuk stuff is |
75 |
> > > > disabled |
76 |
> > > > in system-settings. |
77 |
> > > > I have 200 tasks (yes, nice round figure) |
78 |
> > > > |
79 |
> > > > > Anyone have any ideas? Thanks much. |
80 |
> > > > |
81 |
> > > > For the amount of tasks, check that you are not starting too many |
82 |
> > > > unneeded |
83 |
> > > > services. For the load-average of 1, shouldn't be too much of an |
84 |
> > > > issue, |
85 |
> > > > had |
86 |
> > > > similar in the past with a lot of stuff running and slow disks. |
87 |
> > > > |
88 |
> > > > For the freezing, I would suggest checking all the powersave options, |
89 |
> > > > especially the ones for the harddrives. |
90 |
> > > > Is there anything in the logs when this happens? Eg. check the logs |
91 |
> > > > right |
92 |
> > > > after the system becomes responsible again, maybe there is a hint |
93 |
> > > > there |
94 |
> > > > what is causing this. |
95 |
> > > |
96 |
> > > Unless systemd is setting some powersave options, I certainly never set |
97 |
> > > anything like that, this is a desktop machine, not even a laptop. Next |
98 |
> > > time this happens I will check the logs. Does systemd set some |
99 |
> > > powersave options by default? |
100 |
> > |
101 |
> > I do not know that for sure, best wait for more knowledgable systemd users |
102 |
> > to answer that. If it doesn't, then systemd itself is causing more |
103 |
> > freezes (as per your experience) then openrc. |
104 |
> > |
105 |
> > I would guess it does or at least with the default configuration. What you |
106 |
> > describe makes me think the disks are switched to powersave sooner with |
107 |
> > systemd. |
108 |
> > Can you provide the output of the following command: |
109 |
> > # hdparm -B /dev/sda |
110 |
> > to get the APM settings of the disk. (If you have multiple disks, please |
111 |
> > run it for the others as well. |
112 |
> > |
113 |
> > Question for others as well, how do you get the current setting for the |
114 |
> > spindown timeout set with " hdparm -S <value> <device> "? |
115 |
> > I couldn't find it. |
116 |
> > |
117 |
> > I am happy with openrc and have no intention on switching to systemd as I |
118 |
> > haven't heard of a single feature that would actually make my life easier. |
119 |
> |
120 |
> I don't have hdparm on the system, is it only for older disks? If |
121 |
> memory serves, it did not work at all when I tried it as my disks are |
122 |
> all /dev/sda, etc, but that may be wrong. |
123 |
|
124 |
It also works on new SATA drives and SSDs: |
125 |
|
126 |
# smartctl -a /dev/sda |
127 |
smartctl 6.1 2013-03-16 r3800 [x86_64-linux-3.12.20-gentoo] (local build) |
128 |
Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org |
129 |
|
130 |
=== START OF INFORMATION SECTION === |
131 |
Device Model: INTEL SSDMCEAC120B3 |
132 |
Serial Number: CVLI3223002B120E |
133 |
LU WWN Device Id: 5 5cd2e4 00028738f |
134 |
Firmware Version: LLLi |
135 |
User Capacity: 120,034,123,776 bytes [120 GB] |
136 |
Sector Size: 512 bytes logical/physical |
137 |
Rotation Rate: Solid State Device |
138 |
Device is: Not in smartctl database [for details use: -P showall] |
139 |
ATA Version is: ACS-2 (minor revision not indicated) |
140 |
SATA Version is: SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s) |
141 |
Local Time is: Fri Jun 6 12:46:59 2014 CEST |
142 |
SMART support is: Available - device has SMART capability. |
143 |
SMART support is: Enabled |
144 |
|
145 |
# hdparm /dev/sda |
146 |
|
147 |
/dev/sda: |
148 |
multcount = 16 (on) |
149 |
IO_support = 1 (32-bit) |
150 |
readonly = 0 (off) |
151 |
readahead = 256 (on) |
152 |
geometry = 14593/255/63, sectors = 234441648, start = 0 |
153 |
|
154 |
# hdparm -B /dev/sda |
155 |
|
156 |
/dev/sda: |
157 |
APM_level = 254 |
158 |
|
159 |
There might also be other ways to configure the powermanagement settings, I |
160 |
haven't looked into those yet. |
161 |
|
162 |
-- |
163 |
Joost |