1060195 – Lack of 64-Bit Counters in /sbin/ip -s link show

RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.

Bug 1060195 - Lack of 64-Bit Counters in /sbin/ip -s link show

Summary: Lack of 64-Bit Counters in /sbin/ip -s link show

Keywords:
Status:	CLOSED ERRATA
Alias:	None
Product:	Red Hat Enterprise Linux 6
Classification:	Red Hat
Component:	iproute
Sub Component:
Version:	6.5
Hardware:	x86_64
OS:	Linux
Priority:	unspecified
Severity:	low
Target Milestone:	rc
Target Release:	---
Assignee:	Pavel Šimerda (pavlix)
QA Contact:	Jaroslav Aster
Docs Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+	depends on / blocked

Reported:	2014-01-31 12:37 UTC by Ted Rule
Modified:	2019-07-11 07:51 UTC (History)
CC List:	5 users (show)
Fixed In Version:	iproute-2.6.32-38.el6
Doc Type:	Bug Fix
Doc Text:	The /sbin/ip file displayed statistics for 32-bit system even on 64-bit kernel. To fix this bug, a patch has been applied, and the statistics are now presented correctly.
Clone Of:
Environment:
Last Closed:	2015-07-22 06:40:53 UTC
Target Upstream Version:
Embargoed:

Attachments	(Terms of Use)
patch (12.92 KB, patch) 2014-12-11 20:35 UTC, Pavel Šimerda (pavlix)	no flags	Details \| Diff
patch (13.13 KB, patch) 2015-02-20 12:39 UTC, Pavel Šimerda (pavlix)	no flags	Details \| Diff
Show Obsolete (1) View All

Links
System	ID	Private	Priority	Status	Summary	Last Updated
Red Hat Knowledge Base (Solution)	1159143	0	None	None	None	Never
Red Hat Product Errata	RHBA-2015:1331	0	normal	SHIPPED_LIVE	iproute bug fix and enhancement update	2015-07-20 17:53:02 UTC

Description Ted Rule 2014-01-31 12:37:40 UTC

Description of problem:

/sbin/ip shows 32-bit statistics, even on 64-bit kernel.


Version-Release number of selected component (if applicable):

For this kernel/iproute/net-tools combination on my test rig:

[root@trafalgar-centos65-x86-64 ~]# uname -a 
Linux trafalgar-centos65-x86-64.bugfinder.co.uk 2.6.32-431.el6.x86_64 #1 SMP Fri Nov 22 03:15:09 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
[root@trafalgar-centos65-x86-64 ~]# 

iproute-2.6.32-31.el6.x86_64

net-tools-1.60-110.el6_2.x86_64


How reproducible:

Having pumped about 5GB of data in to a test CentOS6.5 x86_64 instance, I could see that /sbin/ifconfig reported >4GB of Rx Bytes, whilst "/sbin/ip -s link show" reported 600MB, indicating that ifconfig is using a 64-bit counter whilst /sbin/ip is using a 32-bit counter.

FWIW, SNMP interfaces MIB still reports a 32-bit Counter, but that's to be expected as it's defined as 32-bit in the MIB, but the HC-MIB does report 64-bit Counters on x86_64 as shown by the tests below.


[root@trafalgar-centos65-x86-64 ~]# ifconfig eth0
eth0      Link encap:Ethernet  HWaddr 00:1C:42:91:BE:74 
          inet addr:10.211.55.8  Bcast:10.211.55.255  Mask:255.255.255.0
          inet6 addr: fe80::21c:42ff:fe91:be74/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:3455650 errors:0 dropped:0 overruns:0 frame:0
          TX packets:463111 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:4931589231 (4.5 GiB)  TX bytes:33046767 (31.5 MiB)

[root@trafalgar-centos65-x86-64 ~]#


[root@trafalgar-centos65-x86-64 ~]# ip -s link show eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:1c:42:91:be:74 brd ff:ff:ff:ff:ff:ff
    RX: bytes  packets  errors  dropped overrun mcast  
    636623473  3455667  0       0       0       1     
    TX: bytes  packets  errors  dropped carrier collsns
    33048677   463120   0       0       0       0     
[root@trafalgar-centos65-x86-64 ~]#


FWIW, neither ifconfig nor /sbin/ip nor SNMP HC-MIB report 64-bit Counters on 32-bit Kernels, but I understand that part of the reason for that may be difficulties in updating a 64-bit kernel variable atomically on a 32-bit ARCH.

It would be good, in my view, to not support HC-MIB on 32-bit ARCH unless it can report a true 64-bit Counter, but that may be something of a religious argument :)




[root@trafalgar-centos65-x86-64 ~]# snmpwalk -c public -v 2c localhost interfaces | grep Octets
IF-MIB::ifInOctets.1 = Counter32: 0
IF-MIB::ifInOctets.2 = Counter32: 636619429
IF-MIB::ifOutOctets.1 = Counter32: 0
IF-MIB::ifOutOctets.2 = Counter32: 33042059
[root@trafalgar-centos65-x86-64 ~]#

[root@trafalgar-centos65-x86-64 ~]# snmpwalk -c public -v 2c localhost .1.3.6.1.2.1.31  | grep Octets
IF-MIB::ifHCInOctets.1 = Counter64: 13572
IF-MIB::ifHCInOctets.2 = Counter64: 4931600285
IF-MIB::ifHCOutOctets.1 = Counter64: 13572
IF-MIB::ifHCOutOctets.2 = Counter64: 33058003
[root@trafalgar-centos65-x86-64 ~]#

Comment 6 Pavel Šimerda (pavlix) 2014-12-11 20:35:53 UTC

Created attachment 967374 [details]
patch

Comment 9 Jaroslav Aster 2015-02-18 16:42:29 UTC

Hi,

there is a mistake in printing rx/tx stats, the number is correct.

# # rpm -q iproute
iproute-2.6.32-37.el6.x86_64

# ip -s l show dev lo
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    RX: bytes  packets  errors  dropped overrun mcast   
    412714401141825064980      0      0      0      
    TX: bytes  packets  errors  dropped carrier collsns 
    412714401141825064980      0      0      0      

# ifconfig lo
lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:65536  Metric:1
          RX packets:182506498 errors:0 dropped:0 overruns:0 frame:0
          TX packets:182506498 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:41271440114 (38.4 GiB)  TX bytes:41271440114 (38.4 GiB)

You can see that there are no spaces between rx/tx bytes, packages and errors, so there is one number instead of three numbers.

Comment 11 Pavel Šimerda (pavlix) 2015-02-20 12:39:51 UTC

Created attachment 993889 [details]
patch

Also backported the print_num() function that is used in the code.

Comment 16 errata-xmlrpc 2015-07-22 06:40:53 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHBA-2015-1331.html

Note You need to log in before you can comment on or make changes to this bug.