Submitted by fuggi on Thu, 05/16/2013 - 19:47
Hello,
the System Information page states incorrect memory information since the update of Webmin to version 1.630 a few hours ago.
It reads:
Real memory 8589934592 TB total, 152 MB used
Both values are (presumably) wrong, because this particular VPS has 2 GB of RAM and usually uses between 250 and 350 MB according to previous Webmin versions.
Regards,
fuggi
Status:
Closed (fixed)
Comments
Submitted by andreychek on Thu, 05/16/2013 - 20:12 Comment #1
Howdy -- thanks for the bug report! What is the output of the command
free -m
on your system?Submitted by fuggi on Fri, 05/17/2013 - 02:43 Comment #2
Hi,
it reads:
total used free shared buffers cached
Mem: 2048 540 1507 0 0 253
-/+ buffers/cache: 287 1760
Swap: 0 0 0
Regards,
fuggi
EDIT: I just wrote my comment. I did not change any assignments and, thus, do not know why it has been changed.
Submitted by andreychek on Fri, 05/17/2013 - 08:35 Comment #3
Jamie, this sounds suspiciously like a bug :-)
Submitted by fuggi on Fri, 05/17/2013 - 11:09 Comment #4
I thought so. :-)
Submitted by JamieCameron on Sat, 05/18/2013 - 01:04 Comment #5
A couple of people have reported this, on OpenVZ systems.
Did you perhaps upgrade the kernel or other packages at the same time as Webmin?
Submitted by allhen on Sat, 05/18/2013 - 02:41 Comment #6
Same for me !!
Submitted by fuggi on Sat, 05/18/2013 - 09:16 Comment #7
No, only webmin and usermin.
However, I updated linux-libc-dev from 2.6.32-48squeeze1 to 2.6.32-48squeeze3 two days earlier and haven't had a look on the memory information reported by webmin between these two updates.
Regards,
fuggi
Submitted by JamieCameron on Sat, 05/18/2013 - 10:34 Comment #8
I had another look at this, and I think I found the issue - Webmin versions prior to 1.630 weren't properly reading the upper memory limit from the /proc/user_beancounters file due to a parsing error. However the new release does parse that file properly ... and picks up a maximum memory limit that is crazily high.
The solution is to download and install the fix for the Running Processes module from http://www.webmin.com/updates.html
Submitted by allhen on Sat, 05/18/2013 - 11:38 Comment #9
Thanks, problem solved with this update.
Submitted by fuggi on Sat, 05/18/2013 - 14:19 Comment #10
Works! Thank you!
Submitted by fuggi on Sun, 05/19/2013 - 08:13 Comment #11
I'm sorry, but I see the opposite problem on another OpenVZ VPS running Debian 6.0 64 bit (as opposed to 32 bit on the first machine): the total available memory and the used memory are both less:
Real memory 1024 MB total, 237.93 MB used
The output of
free -m
reads:total used free shared buffers cached
Mem: 2048 935 1112 0 0 0
-/+ buffers/cache: 935 1112
Swap: 2048 0 2048
Installing the fix did not help.
Regards,
fuggi
Submitted by JamieCameron on Sun, 05/19/2013 - 12:39 Comment #12
@fuggi - what does the
/proc/user_beancounters
file contain on this system?Also, what does
/proc/meminfo
contain?Submitted by fuggi on Sun, 05/19/2013 - 13:10 Comment #13
/proc/user_beancounters
:Version: 2.5
uid resource held maxheld barrier limit failcnt
75827: kmemsize 14182234 14210991 96636764 107374182 0
lockedpages 0 0 2059 2059 0
privvmpages 242090 242209 1048576 1101004 0
shmpages 952 952 65536 65536 0
dummy 0 0 9223372036854775807 9223372036854775807 0
numproc 62 62 500 500 0
physpages 141911 142004 0 9223372036854775807 0
vmguarpages 0 0 524288 9223372036854775807 0
oomguarpages 141911 142004 9223372036854775807 9223372036854775807 0
numtcpsock 21 21 550 550 0
numflock 8 8 1000 1100 0
numpty 1 1 102 102 0
numsiginfo 0 0 1024 1024 0
tcpsndbuf 388096 388096 5280000 7392000 0
tcprcvbuf 278528 278528 5280000 7392000 727
othersockbuf 260032 260032 3840000 5376000 0
dgramrcvbuf 0 0 3584000 3584000 0
numothersock 146 146 400 400 0
dcachesize 1094874 1101447 14495514 16106127 0
numfile 2986 2987 17600 17600 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
numiptent 38 38 9223372036854775807 9223372036854775807 0
/proc/meminfo
:MemTotal: 2097152 kB
MemFree: 1128872 kB
Buffers: 0 kB
Cached: 0 kB
SwapCached: 0 kB
Active: 0 kB
Inactive: 0 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 2097152 kB
LowFree: 1128872 kB
SwapTotal: 2097152 kB
SwapFree: 2097152 kB
Dirty: 0 kB
Writeback: 0 kB
AnonPages: 0 kB
Mapped: 0 kB
Slab: 0 kB
PageTables: 0 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
CommitLimit: 0 kB
Committed_AS: 0 kB
VmallocTotal: 0 kB
VmallocUsed: 0 kB
VmallocChunk: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
Hugepagesize: 2048 kB
Submitted by JamieCameron on Sun, 05/19/2013 - 17:56 Comment #14
Interesting, it looks like
/proc/user_beancounters
says that the memory limit is 1 GB (from theprivvmpages
line). I'd be interested to know what your VPS provider has set as the guaranteed and burstable limits for this VM.Submitted by fuggi on Mon, 05/20/2013 - 12:57 Comment #15
I'm sorry, I simply do not know what "burstable" means. Could you explain, please?
Anyway, I know that the guaranteed memory equals the maximum memory and is 2 GB by contract. Do you think I should ask the provider to fix this issue?
Regards,
fuggi
EDIT: Does Virtualmin obtain the value of memory in use from /proc/user_beancounters either?
Submitted by bentrafford on Mon, 05/20/2013 - 11:38 Comment #16
Same issue here. Virtualmin reports my real memory as:
8589934592 TB total, 416 MB used
/proc/user_beancounters
Version: 2.5
uid resource held maxheld barrier limit failcnt
1835: kmemsize 55930861 56872960 9223372036854775807 9223372036854775807 0
lockedpages 0 0 9223372036854775807 9223372036854775807 0
privvmpages 418534 435832 9223372036854775807 9223372036854775807 0
shmpages 8337 8353 9223372036854775807 9223372036854775807 0
dummy 0 0 9223372036854775807 9223372036854775807 0
numproc 126 147 9223372036854775807 9223372036854775807 0
physpages 395276 397550 1048576 1048576 0
vmguarpages 0 0 9223372036854775807 9223372036854775807 0
oomguarpages 248552 248552 9223372036854775807 9223372036854775807 0
numtcpsock 47 48 9223372036854775807 9223372036854775807 0
numflock 34 34 9223372036854775807 9223372036854775807 0
numpty 1 1 9223372036854775807 9223372036854775807 0
numsiginfo 0 30 9223372036854775807 9223372036854775807 0
tcpsndbuf 920816 938256 9223372036854775807 9223372036854775807 0
tcprcvbuf 770048 786432 9223372036854775807 9223372036854775807 0
othersockbuf 402520 409952 9223372036854775807 9223372036854775807 0
dgramrcvbuf 0 4360 9223372036854775807 9223372036854775807 0
numothersock 384 386 9223372036854775807 9223372036854775807 0
dcachesize 23033735 23120953 9223372036854775807 9223372036854775807 0
numfile 6387 6435 9223372036854775807 9223372036854775807 0
dummy 0 0 9223372036854775807 9223372036854775807 0
dummy 0 0 9223372036854775807 9223372036854775807 0
dummy 0 0 9223372036854775807 9223372036854775807 0
numiptent 52 52 9223372036854775807 9223372036854775807 0
/proc/meminfo
MemTotal: 4194304 kB
MemFree: 2645484 kB
Cached: 532204 kB
Active: 789196 kB
Inactive: 706776 kB
Active(anon): 405488 kB
Inactive(anon): 558280 kB
Active(file): 383708 kB
Inactive(file): 148496 kB
Unevictable: 0 kB
Mlocked: 0 kB
SwapTotal: 0 kB
SwapFree: 0 kB
Dirty: 120 kB
Writeback: 0 kB
AnonPages: 963768 kB
Shmem: 33348 kB
Slab: 52816 kB
SReclaimable: 22500 kB
SUnreclaim: 30316 kB
I installed the 1.630 Running Processes module to no avail. This is on a virtual server running CentOS 6.4.
Submitted by JamieCameron on Mon, 05/20/2013 - 17:58 Comment #17
Burstable memory is the limit to which your VM can use up to, but only for a limited period. OpenVZ supports both guaranteed and burstable limits - it is possible that in your case, the guaranteed limit is only 1GB.
Personally I dislike the use of burstable limits, because they cause problems if memory use remains high for a long period.
Submitted by fuggi on Mon, 05/20/2013 - 18:50 Comment #18
Thank you for your explanation! However, as I wrote in my previous post, the contractually guaranteed limit is 2 GB exactly like the burstable limit. I asked the provider to fix this problem and get back to you as soon as I will have news.
Regards,
fuggi
I have this issue on the GPL version. Started yesterday after updating two Virtualmin modules in the last 48 hours. These are the modules. I had posted this yesterday in the general forum:
Here is what free - m
Virtualmin Info: 1024 MB total, 544.67 MB used
OpenVZ and CentOS Linux 6.4
Submitted by qbattersby on Tue, 05/21/2013 - 11:26 Comment #20
Hi,
I am also now having the same issue after updating to webmin 1.630-1
Virtualmin reads out at 1/4 of what it is in actuality (approx). I have 2048MB, it shows 512MB.
Here is the output of free -m
total used free shared buffers cached
Mem: 2048 399 1648 0 0 0
-/+ buffers/cache: 399 1648
Swap: 0 0 0
Sorry should note, on OVZ, CentOS 6. Let me know if you need anything from me to help resolve.
Submitted by JamieCameron on Tue, 05/21/2013 - 13:20 Comment #21
@qbattersby - I would also be interested in seeing the
privvmpages
line from your/proc/user_beancounters
file.Submitted by qbattersby on Tue, 05/21/2013 - 18:45 Comment #22
Here you are;
Version: 2.5
uid resource held maxheld barrier limit failcnt
22663: kmemsize 15566339 72232285 2147483646 2147483646 0
lockedpages 0 20736 999999 999999 0
privvmpages 104778 470335 524288 524288 0
shmpages 1058 140226 262144 262144 2
dummy 0 0 0 0 0
numproc 263 500 500 500 5771
physpages 57166 356751 0 9223372036854775807 0
vmguarpages 0 0 262144 262144 0
oomguarpages 57167 356752 262144 262144 0
numtcpsock 33 1730 7999992 7999992 0
numflock 5 58 999999 999999 0
numpty 1 2 500000 500000 0
numsiginfo 0 99 999999 999999 0
tcpsndbuf 584336 26847880 214748160 396774400 0
tcprcvbuf 540672 26901736 214748160 396774400 0
othersockbuf 208728 746032 214748160 396774400 0
dgramrcvbuf 0 63240 214748160 396774400 0
numothersock 130 436 7999992 7999992 0
dcachesize 727725 2622627 2147483646 2147483646 0
numfile 5436 12071 23999976 23999976 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
numiptent 41 41 999999 999999 0
Submitted by nuganics on Tue, 05/21/2013 - 22:36 Comment #23
Have have the same issue. Sorry if meant to open a new issue. I get "Real memory 8589934592 TB total, 670 MB used" in virtualmin. Have 2560MB RAM with 5120 MB Burst RAM. Virtualmin version 3.99.gpl GPL, Operating system CentOS Linux 6.4 x64 on OpenVZ, Webmin version 1.630, Kernel and CPU Linux 2.6.32-042stab076.5 on x86_64.
[root@server1 ~]# free -m
total used free shared buffers cached
Mem: 2560 1667 892 0 0 787
-/+ buffers/cache: 879 1680
Swap: 5120 197 4922
[root@server1 ~]#
Submitted by andreychek on Tue, 05/21/2013 - 23:40 Comment #24
nuganics, can you post the contents of your /proc/user_beancounters file? Thanks!
Submitted by JamieCameron on Wed, 05/22/2013 - 00:02 Comment #25
Thanks folks - it looks like on some systems, the OpenVZ memory page size is not always the 1k Webmin expects. We will include a fix for this in the next release.
Submitted by nuganics on Wed, 05/22/2013 - 00:12 Comment #26
I basically have no understand of unix so get back to me if the below is wrong. :)
[root@server1 proc]# cat user_beancounters Version: 2.5 uid resource held maxheld barrier limit failcnt 10472: kmemsize 138198538 170876928 922337203685 4775807 9223372036854775807 0 lockedpages 0 0 922337203685 4775807 9223372036854775807 0 privvmpages 555920 803692 922337203685 4775807 9223372036854775807 0 shmpages 1829 2533 922337203685 4775807 9223372036854775807 0 dummy 0 0 0 0 0 numproc 177 357 922337203685 4775807 9223372036854775807 0 physpages 484812 655237 0 655360 0 vmguarpages 0 0 922337203685 4775807 9223372036854775807 0 oomguarpages 238122 391231 922337203685 4775807 9223372036854775807 0 numtcpsock 187 226 922337203685 4775807 9223372036854775807 0 numflock 179 186 922337203685 4775807 9223372036854775807 0 numpty 1 1 922337203685 4775807 9223372036854775807 0 numsiginfo 15 75 922337203685 4775807 9223372036854775807 0 tcpsndbuf 3636504 4305632 922337203685 4775807 9223372036854775807 0 tcprcvbuf 3063808 3752856 922337203685 4775807 9223372036854775807 0 othersockbuf 298136 1650664 922337203685 4775807 9223372036854775807 0 dgramrcvbuf 0 64736 922337203685 4775807 9223372036854775807 0 numothersock 333 468 922337203685 4775807 9223372036854775807 0 dcachesize 86469144 86735109 67 1088640 671088640 0 numfile 5957 10180 922337203685 4775807 9223372036854775807 0 dummy 0 0 0 0 0 dummy 0 0 0 0 0 dummy 0 0 0 0 0 numiptent 41 41 922337203685 4775807 9223372036854775807 0 [root@server1 proc]#
Submitted by nuganics on Wed, 05/22/2013 - 00:13 Comment #27
above at http://pastebin.com/ar0uJXEh
Submitted by JamieCameron on Wed, 05/22/2013 - 15:27 Comment #28
nuganics - the fix from http://www.webmin.com/updates.html should solve your issue.
Submitted by qbattersby on Wed, 05/22/2013 - 19:11 Comment #29
Hey I tried this fix with no luck. :(
Submitted by nuganics on Wed, 05/22/2013 - 19:34 Comment #30
worked for me after the update install. In Virtualmin at top I clicked "refresh system information" and shows correctly. Thanks guys.
Submitted by qbattersby on Wed, 05/22/2013 - 19:42 Comment #31
Yea tried all of that and did a reboot, no love. :(
Submitted by JamieCameron on Wed, 05/22/2013 - 23:19 Comment #32
@qbattersby - I've released a further update at http://www.webmin.com/updates.html that should fix your issue.
Submitted by qbattersby on Thu, 05/23/2013 - 07:13 Comment #33
Thanks Jamie - all fixed now!
Submitted by fuggi on Fri, 05/24/2013 - 07:15 Comment #34
Hi Jamie,
according to my provider (it took him 4 days to answer), privvmpages in user_beancounters gives the size of the burstable memory in number of 4k pages. Therefore, I have got 2 GB guaranteed memory, but 4 GB burstable memory (what I didn't know before). Your latest update fixes this issue as "Real memory" on the "System Information" page now reads "4 GB total, 1 GB used".
However, IMHO I would prefer a distinction between guaranteed and burstable memory on the "System Information" page. What do you think?
Regards,
fuggi
Submitted by JamieCameron on Fri, 05/24/2013 - 12:31 Comment #35
That's a good idea .. I will look into displaying the guaranteed memory separately.
Submitted by fuggi on Fri, 05/24/2013 - 15:23 Comment #36
I'm glad you like my idea and looking forward to seeing it implemented in the next release! ;)
Hi,
== SPECS ==
OS: CentOS 5/6
Webmin: 1.630
Virtualmin: 3.99.gpl
Platform: OpenVZ
As noted above for others, memory is being read by Virtualmin as half the actual guaranteed memory.
-Peter
Submitted by fuggi on Tue, 05/28/2013 - 06:12 Comment #38
Hi Peter,
have you installed the fix from http://www.webmin.com/updates.html?
Regards,
fuggi
Fuggi,
I've got 1.630 installed if that's what you're asking...
-Peter
Submitted by fuggi on Wed, 05/29/2013 - 05:54 Comment #40
Peter,
sorry, no, it isn't! Please have a look on the web page I mentioned above and install the one and only fix listed there for webmin version 1.630 on your server via "Webmin" -> "Webmin Configuration" -> "Webmin Modules".
Regards,
fuggi
Fuggi,
Thanks for the clarification, all is working once again :-)
-Peter
Submitted by fuggi on Fri, 05/31/2013 - 04:40 Comment #42
Peter,
please read carefully next time before you re-open an issue.
Regards,
fuggi
Submitted by andreychek on Fri, 05/31/2013 - 07:48 Comment #43
It's okay, even I didn't understand that at first. We appreciate that Peter is trying to make sure it's working properly.
Thanks to everyone for their input on this issue!