Smartctl running at 100% CPU after only 20mins

M

micle546

Cadet
Joined
May 28, 2021
Messages
6
So, after only 20 mins of system uptime, my CPUs are pegged at 100%. I believe it's got something to do with the Raid card consuming the SMART data at this point.

This error repeats in the terminal, over and over again, every couple of minutes. I'm assuming something is getting hung, and the process keeps running without terminating, but I've no idea there.
Code:
May 28 15:03:33 truenas 1 2021-05-28T15:03:33.399886-04:00 truenas.home.local collectd 1674 - - Traceback (most recent call last):
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/python3.8/site-packages/middlewared/client/client.py", line 429, in call
    raise ClientException(c.error, c.errno, c.trace, c.extra)
middlewared.client.client.ClientException: Lock is not acquired.
May 28 15:08:33 truenas 1 2021-05-28T15:08:33.422123-04:00 truenas.home.local collectd 1674 - - Traceback (most recent call last):
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/python3.8/site-packages/middlewared/client/client.py", line 429, in call
    raise ClientException(c.error, c.errno, c.trace, c.extra)
middlewared.client.client.ClientException: Lock is not acquired.


The disks have never once reported temperatures. I've got a mix of old and new drives in there, (The new ones are IronWolf NAS Pro 4TB drives)

simple top output:
Code:
last pid:  2942;  load averages: 13.14, 11.87,  9.10                                                                                                                                          up 0+00:46:56  15:22:21
61 processes:  14 running, 47 sleeping
CPU: 14.3% user,  0.0% nice, 85.7% system,  0.0% interrupt,  0.0% idle
Mem: 854M Active, 675M Inact, 33M Laundry, 28G Wired, 1528M Free
ARC: 26G Total, 69M MFU, 25G MRU, 120M Anon, 67M Header, 207M Other
     25G Compressed, 30G Uncompressed, 1.24:1 Ratio
Swap: 6144M Total, 6144M Free

  PID USERNAME    THR PRI NICE   SIZE    RES STATE    C   TIME    WCPU COMMAND
 2815 root          1  81    0    15M  5224K RUN      1   3:01  33.15% smartctl
 2892 root          1  81    0    15M  5224K RUN      1   1:08  32.20% smartctl
 2891 root          1  81    0    15M  5232K RUN      2   1:08  30.88% smartctl
 2750 root          1  81    0    15M  5232K RUN      1   5:27  30.80% smartctl
 2816 root          1  81    0    15M  5232K CPU3     3   3:00  30.69% smartctl
 2692 root          1  81    0    15M  5232K RUN      2   8:50  29.48% smartctl
 2626 root          1  81    0    15M  5224K RUN      2  12:55  29.27% smartctl
 2880 root          1  81    0    15M  5232K RUN      2   1:13  29.17% smartctl
 2804 root          1  81    0    15M  5224K RUN      2   3:07  29.05% smartctl
 2679 root          1  81    0    15M  5224K CPU2     2   9:01  28.83% smartctl
 2624 root          1  81    0    15M  5232K RUN      1  12:57  28.31% smartctl
 2749 root          1  81    0    15M  5224K CPU1     1   5:27  28.14% smartctl
  273 root         45  52    0   625M   498M kqread   1   1:41  15.11% python3.8
 2048 root          1  79    0   189M   168M RUN      0  17:06  12.37% smbd
  287 root          3  20    0   199M   160M usem     0   0:07   1.16% python3.8
 1956 root          1  20    0   170M   151M kqread   0   0:06   0.14% smbd
 2932 root          1  20    0    13M  3984K CPU0     0   0:00   0.09% top
 2130 root          1  20    0    11M  2820K pause    2   0:00   0.01% iostat
  256 root          1  20    0    11M  1964K select   3   0:00   0.01% devd
 2504 root          1  20    0    28M    14M select   1   0:00   0.01% sshd
 1649 www           1  20    0    35M  9144K kqread   0   0:00   0.01% nginx
 1552 ntpd          1  20    0    19M  6864K select   3   0:00   0.01% ntpd
 1639 root          8  20    0    36M    16M select   3   0:03   0.00% rrdcached
  259 root          1  20    0    17M  6360K select   2   0:00   0.00% zfsd
 1940 root          1  20    0    88M    70M kqread   1   0:00   0.00% winbindd
 1674 root         11  20    0    85M    44M nanslp   3   0:37   0.00% collectd
  288 root          3  20    0   211M   161M usem     0   0:07   0.00% python3.8
  285 root          3  20    0   199M   160M piperd   2   0:07   0.00% python3.8
  286 root          3  20    0   199M   160M usem     0   0:06   0.00% python3.8
  284 root          3  20    0   199M   160M usem     0   0:06   0.00% python3.8
 1680 avahi         1  20    0    13M  4324K select   3   0:01   0.00% avahi-daemon
  432 root          1  20    0    58M    46M zevent   3   0:01   0.00% python3.8
 1831 root          1  52    0    48M    35M ttyin    0   0:00   0.00% python3.8
 1931 root          1  21    0   172M   153M kqread   3   0:00   0.00% smbd
 1947 daemon        3  20    0    71M    48M kqread   2   0:00   0.00% python3.8

12 smartctl tasks =?= 12 disks in the pool?

TrueNAS has never been able to figure out what the disks were, but it can pick up serial#, the model# and size, but none of the smart stuff comes through.

If I had to guess that's the fault of the raid card, even though it's in JBOD mode. I tried to replace it with a newer card, but the backplane doesn't seem to like my new SAS3 card (LSI SAS9340-8i ServeRAID M1215). Isn't SAS backwards compatible?

CPU: Dual Intel(R) Xeon(R) CPU 5160 @ 3Ghz
Mobo: X7DB3
Ram: 8x2Gb
Controller: Acrea arc-1880i in JBOD mode
Backplane: bpn-sas-836el1
 
Spearfoot

Spearfoot

He of the long foot
Moderator
Joined
May 13, 2015
Messages
2,478
micle546 said:
So, after only 20 mins of system uptime, my CPUs are pegged at 100%. I believe it's got something to do with the Raid card consuming the SMART data at this point.

This error repeats in the terminal, over and over again, every couple of minutes. I'm assuming something is getting hung, and the process keeps running without terminating, but I've no idea there.
Code:
May 28 15:03:33 truenas 1 2021-05-28T15:03:33.399886-04:00 truenas.home.local collectd 1674 - - Traceback (most recent call last):
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/python3.8/site-packages/middlewared/client/client.py", line 429, in call
    raise ClientException(c.error, c.errno, c.trace, c.extra)
middlewared.client.client.ClientException: Lock is not acquired.
May 28 15:08:33 truenas 1 2021-05-28T15:08:33.422123-04:00 truenas.home.local collectd 1674 - - Traceback (most recent call last):
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/collectd_pyplugins/disktemp.py", line 63, in read
    temperatures = c.call('disk.temperatures', self.disks, self.powermode)
  File "/usr/local/lib/python3.8/site-packages/middlewared/client/client.py", line 429, in call
    raise ClientException(c.error, c.errno, c.trace, c.extra)
middlewared.client.client.ClientException: Lock is not acquired.


The disks have never once reported temperatures. I've got a mix of old and new drives in there, (The new ones are IronWolf NAS Pro 4TB drives)

simple top output:
Code:
last pid:  2942;  load averages: 13.14, 11.87,  9.10                                                                                                                                          up 0+00:46:56  15:22:21
61 processes:  14 running, 47 sleeping
CPU: 14.3% user,  0.0% nice, 85.7% system,  0.0% interrupt,  0.0% idle
Mem: 854M Active, 675M Inact, 33M Laundry, 28G Wired, 1528M Free
ARC: 26G Total, 69M MFU, 25G MRU, 120M Anon, 67M Header, 207M Other
     25G Compressed, 30G Uncompressed, 1.24:1 Ratio
Swap: 6144M Total, 6144M Free

  PID USERNAME    THR PRI NICE   SIZE    RES STATE    C   TIME    WCPU COMMAND
2815 root          1  81    0    15M  5224K RUN      1   3:01  33.15% smartctl
2892 root          1  81    0    15M  5224K RUN      1   1:08  32.20% smartctl
2891 root          1  81    0    15M  5232K RUN      2   1:08  30.88% smartctl
2750 root          1  81    0    15M  5232K RUN      1   5:27  30.80% smartctl
2816 root          1  81    0    15M  5232K CPU3     3   3:00  30.69% smartctl
2692 root          1  81    0    15M  5232K RUN      2   8:50  29.48% smartctl
2626 root          1  81    0    15M  5224K RUN      2  12:55  29.27% smartctl
2880 root          1  81    0    15M  5232K RUN      2   1:13  29.17% smartctl
2804 root          1  81    0    15M  5224K RUN      2   3:07  29.05% smartctl
2679 root          1  81    0    15M  5224K CPU2     2   9:01  28.83% smartctl
2624 root          1  81    0    15M  5232K RUN      1  12:57  28.31% smartctl
2749 root          1  81    0    15M  5224K CPU1     1   5:27  28.14% smartctl
  273 root         45  52    0   625M   498M kqread   1   1:41  15.11% python3.8
2048 root          1  79    0   189M   168M RUN      0  17:06  12.37% smbd
  287 root          3  20    0   199M   160M usem     0   0:07   1.16% python3.8
1956 root          1  20    0   170M   151M kqread   0   0:06   0.14% smbd
2932 root          1  20    0    13M  3984K CPU0     0   0:00   0.09% top
2130 root          1  20    0    11M  2820K pause    2   0:00   0.01% iostat
  256 root          1  20    0    11M  1964K select   3   0:00   0.01% devd
2504 root          1  20    0    28M    14M select   1   0:00   0.01% sshd
1649 www           1  20    0    35M  9144K kqread   0   0:00   0.01% nginx
1552 ntpd          1  20    0    19M  6864K select   3   0:00   0.01% ntpd
1639 root          8  20    0    36M    16M select   3   0:03   0.00% rrdcached
  259 root          1  20    0    17M  6360K select   2   0:00   0.00% zfsd
1940 root          1  20    0    88M    70M kqread   1   0:00   0.00% winbindd
1674 root         11  20    0    85M    44M nanslp   3   0:37   0.00% collectd
  288 root          3  20    0   211M   161M usem     0   0:07   0.00% python3.8
  285 root          3  20    0   199M   160M piperd   2   0:07   0.00% python3.8
  286 root          3  20    0   199M   160M usem     0   0:06   0.00% python3.8
  284 root          3  20    0   199M   160M usem     0   0:06   0.00% python3.8
1680 avahi         1  20    0    13M  4324K select   3   0:01   0.00% avahi-daemon
  432 root          1  20    0    58M    46M zevent   3   0:01   0.00% python3.8
1831 root          1  52    0    48M    35M ttyin    0   0:00   0.00% python3.8
1931 root          1  21    0   172M   153M kqread   3   0:00   0.00% smbd
1947 daemon        3  20    0    71M    48M kqread   2   0:00   0.00% python3.8

12 smartctl tasks =?= 12 disks in the pool?

TrueNAS has never been able to figure out what the disks were, but it can pick up serial#, the model# and size, but none of the smart stuff comes through.

If I had to guess that's the fault of the raid card, even though it's in JBOD mode. I tried to replace it with a newer card, but the backplane doesn't seem to like my new SAS3 card (LSI SAS9340-8i ServeRAID M1215). Isn't SAS backwards compatible?

CPU: Dual Intel(R) Xeon(R) CPU 5160 @ 3Ghz
Mobo: X7DB3
Ram: 8x2Gb
Controller: Acrea arc-1880i in JBOD mode
Backplane: bpn-sas-836el1
Click to expand...
Are you using the LSI SAS9340-8i? Or the Areca ARC-1880i? You mention both...

If you're using the LSI card, it sounds like you're running it in IR mode -- you said it was 'in JBOD mode'. If this is the case, you're much better off flashing the card to IT mode, as this will give FreeNAS/TrueNAS direct access to your hard drives. FreeNAS/TrueNAS works much better with Host Bus Adapters (HBAs) than with RAID cards.
 
M

micle546

Cadet
Joined
May 28, 2021
Messages
6
I'm using the Acrea 1880i in JBOD mode. I bought the LSI card to replace it, preflashed to IT mode by ArtOfServer, but it doesn't detect the drives. I'm debating on returning the LSI card for an older one
 
M

micle546

Cadet
Joined
May 28, 2021
Messages
6
htop:
1622235171466.png


Smartctl output on one of the drives:
Code:
root@truenas[/dev]# smartctl -a dev/arcmsr0 -d areca,1/2 -a
smartctl 7.2 2020-12-30 r5155 [FreeBSD 12.2-RELEASE-p6 amd64] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

do_scsi_cmnd_io with write buffer failed code = ffffffff
do_scsi_cmnd_io with write buffer failed code = ffffffff
do_scsi_cmnd_io with write buffer failed code = ffffffff
Smartctl open device: dev/arcmsr0 [areca_disk#01_enc#02] failed: Input/output error
root@truenas[/dev]#
 
Spearfoot

Spearfoot

He of the long foot
Moderator
Joined
May 13, 2015
Messages
2,478
micle546 said:
I'm using the Acrea 1880i in JBOD mode. I bought the LSI card to replace it, preflashed to IT mode by ArtOfServer, but it doesn't detect the drives. I'm debating on returning the LSI card for an older one
Click to expand...
Odd. He seems to be a knowledgeable guy and a reputable eBay seller, plus, I'd take the LSI card over the Areca any day of the week.

Your backplane may be the problem: it's a BPN-SAS-836EL1, correct?

The older, SAS1 versions have problems with disks > 2TB in size -- you really want a BPN-SAS2-836EL1 expander backplane.
 
M

micle546

Cadet
Joined
May 28, 2021
Messages
6
Yeah, That's what he suggested. The only problem I see is the difference in mounting hole pattern, not all of the holes in the supermicro bpn-sas-836el1 are screw holes, some of them seem to be shouldered pins, and I don't think it would mount correctly, If I could find a way to mount the sas2 backplane I think it 'could' work, as I 'think' the drive trays are the same, so the drives would connect. See pics

BPN-SAS-836EL1
1622240302748.png


BPN-SAS2-836EL1
1622240281620.png
 
Spearfoot

Spearfoot

He of the long foot
Moderator
Joined
May 13, 2015
Messages
2,478
micle546 said:
Yeah, That's what he suggested. The only problem I see is the difference in mounting hole pattern, not all of the holes in the supermicro bpn-sas-836el1 are screw holes, some of them seem to be shouldered pins, and I don't think it would mount correctly, If I could find a way to mount the sas2 backplane I think it 'could' work, as I 'think' the drive trays are the same, so the drives would connect. See pics

BPN-SAS-836EL1
View attachment 47435

BPN-SAS2-836EL1
View attachment 47434
Click to expand...
I dunno... I've never swapped out a Supermicro backplane.

Why don't you ask Supermicro support? They're usually responsive, and they ought to know: support@supermicro.com

My hunch is that you could make it work -- the form factors are nearly identical.
 
M

micle546

Cadet
Joined
May 28, 2021
Messages
6
They weren't all too helpful. They more or less said: "We don’t sell backplane as individual component, it comes with chassis. Repurposed the older chassis with newer backplane is not easy since newer backplane has different layout and components."
 
Spearfoot

Spearfoot

He of the long foot
Moderator
Joined
May 13, 2015
Messages
2,478
micle546 said:
They weren't all too helpful. They more or less said: "We don’t sell backplane as individual component, it comes with chassis. Repurposed the older chassis with newer backplane is not easy since newer backplane has different layout and components."
Click to expand...
Well... at least they were responsive! Though I think they fibbed a little about not selling backplanes as components; their authorized resellers stock backplanes, wiredzone.com, for example:

www.wiredzone.com

Supermicro BPN-SAS2-836EL1 3U Backplane w/ single LSI SAS2 x28

3U, Backplane w/ single LSI SAS2 x28 Expander Chip, for SC836's Chassis
www.wiredzone.com www.wiredzone.com

Maybe you could ask the folks at wiredzone.com about swapping out your backplane.

An alternative, of course, is to replace your chassis.
 
M

micle546

Cadet
Joined
May 28, 2021
Messages
6
Def a bug though. smartctl never terminates, if it hangs it seems. This is the same server a few hours later.
1622246040190.png
 
You must log in or register to reply here.

Similar threads

A
SOLVED CPU Thread 0 running at 100% ALL THE TIME
Replies
12
Views
3K
Azsumbre13
A
H
  • Locked
FreeNAS 9.2.1 samba conversion error
Replies
16
Views
4K
Hi-Liter
H
G
  • Locked
sshd high CPU occupancy
Replies
6
Views
2K
Ericloewe
Ericloewe
T
  • Locked
High CPU usage during network transfers and network freezes
Replies
7
Views
4K
Touche
T
D
strange RAM usage
2
Replies
21
Views
6K
Dunuin
D
Top