Bug 1108805 - 32-bit limits on iperf testing of High Speed interfaces
Summary: 32-bit limits on iperf testing of High Speed interfaces
Keywords:
Status: CLOSED EOL
Alias: None
Product: Fedora EPEL
Classification: Fedora
Component: iperf
Version: el6
Hardware: All
OS: Linux
unspecified
low
Target Milestone: ---
Assignee: Gabriel Somlo
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2014-06-12 15:28 UTC by Ted Rule
Modified: 2020-11-30 15:06 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2020-11-30 15:06:46 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)

Description Ted Rule 2014-06-12 15:28:47 UTC
Using iperf-2.0.5-11.el6.x86_64:

Using moderately high-speed interfaces, e.g. 1Gigabit, it is possible to overflow the 32-bit packet counter within iperf causing the test to "prematurely" finish.

At 1Gbps, using full-sized UDP packets, it's possible to reach a 2^31 signed 32-bit Integer counter wrap within about 6 hours. This makes it difficult to use iperf to soak test a system for an extended period, even more so if the interface runs at 10Gbps or more.

Looking at the source code, it's apparent that recent code changes have partially fixed this issue by ensuring that several variables, including Total Byte Counts, have been modified to use 64Bit unsigned integers, but the problem still remains for the Total Packet Count and potentially some other counters.

The code logic appears to use a signed 32bit integer as both a Packet Count and a Packet ID which is embedded within the Packet, with the special case of ID == -1 used to signal "end of test".

Ideally, I would imagine that the code should use a 64-bit unsigned integer for the Packet Count/ID, with a separate flag indicating "end-of-test".

To reproduce the problem, run a 1Gbps UDP test between two iperfs for at least 7 hours ==~ 25000 seconds. ( or run for longer at slower speeds to trigger the same 32-bit wrap ).

FWIW, iperf3 seems to have the same issue, but I'm not using that at present due to the lack of multicast support.

Comment 1 Ben Cotton 2020-11-05 16:47:42 UTC
This message is a reminder that EPEL 6 is nearing its end of life. Fedora will stop maintaining and issuing updates for EPEL 6 on 2020-11-30. It is our policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a 'version' of 'el6'.

Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later EPEL version.

Thank you for reporting this issue and we are sorry that we were not able to fix it before EPEL 6 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged  change the 'version' to a later Fedora version prior this bug is closed as described in the policy above.

Comment 2 Ben Cotton 2020-11-30 15:06:46 UTC
EPEL el6 changed to end-of-life (EOL) status on 2020-11-30. EPEL el6 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
EPEL please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.


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