diff options
Diffstat (limited to 'doc/trex_faq.asciidoc')
| -rw-r--r-- | doc/trex_faq.asciidoc | 43 |
1 files changed, 41 insertions, 2 deletions
diff --git a/doc/trex_faq.asciidoc b/doc/trex_faq.asciidoc index 6558ba38..55e83f4e 100644 --- a/doc/trex_faq.asciidoc +++ b/doc/trex_faq.asciidoc @@ -85,10 +85,10 @@ Not yet. Support for FM10K and Mellanox Connectx5 is under development. ==== Is there a GUI? TRex team is not developing it. -You can find GUI for statless mode developed by Exalt company in +You can find GUI for statless mode developed by Exalt/XORED company in link:https://github.com/exalt-tech/trex-stateless-gui#trex-stateless-gui-beta[this link]. + There is also work in progress for packet editor (not released yet) in link:https://github.com/cisco-system-traffic-generator/trex-packet-editor[here]. + -New stateless GUI features are developed in link:https://github.com/cisco-system-traffic-generator/trex-stateless-gui[here]. This is also work in progress, not released yet. +New stateless GUI features are developed in link:https://github.com/cisco-system-traffic-generator/trex-stateless-gui[here]. v3.1.1 was released ==== What is the maximum number of ports per TRex application? 12 ports @@ -226,6 +226,45 @@ Note: you might need additional Linux packages for that: ** sudo apt install build-essential +==== Running with ESXi/KVM I get WATCHDOG crash + +While running TRex on ESXi/KVM I get this error + +[source,bash] +---- +terminate called after throwing an instance of 'std::runtime_error' + what(): WATCHDOG: task 'ZMQ sync request-response' has not responded for more than 1.58321 seconds - timeout is 1 seconds + +*** traceback follows *** + ./_t-rex-64-o() [0x48aa96] + /lib/x86_64-linux-gnu/libpthread.so.0(+0x10340) + __poll + 45 + /home/trex/v2.05/libzmq.so.3(+0x26642) + /home/trex/v2.05/libzmq.so.3(+0x183da) + /home/trex/v2.05/libzmq.so.3(+0x274c3) + /home/trex/v2.05/libzmq.so.3(+0x27b95) + /home/trex/v2.05/libzmq.so.3(+0x3afe9) + TrexRpcServerReqRes::fetch_one_request(std::string&) + TrexRpcServerReqRes::_rpc_thread_cb() + 1062 + /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0xb1a40) + /lib/x86_64-linux-gnu/libpthread.so.0(+0x8182) + clone + 109 +---- + +Try to disable Hypervisor Hyper thread (HT) for TRex threads . two TRex threads can't share cores. +Moving to 'none' (ESXi) only disable sharing of hyper threading, this mitigates the problem but does not solve it. +The full solution is to provide pinning of cores *exclusively* to TRex. +This means making sure on your Hypervisor that *only* TRex VM gets those specific cores and no other guest shares them. +(we handle Linux processor affinity but cannot handle the Hypervisor affinity) +If you cannot make sure your cores are pinned to the TRex guest you can run TRex without watchdog: + +[source,bash] +---- +$sudo ./t-rex-64 -i --no-watchdog +---- + +This will hide the problem that some threads do not get reasonable response time, but will allow to work anyway. + === Stateful |