>No, it's not that simple
Can I update that sub-buffer commit_cold counter(how about hot one?) to the
"right" value then skip consuming this sub-buffer(but update the consume
counter as if I have consumed the sub-buffer)?
It seems the tracing progress will continue with the cost of missing all the
spans in that sub-buffer.
Thanks
zhenyu.ren
------------------------------------------------------------------
发件人：Mathieu Desnoyers <mathieu.desnoy...@efficios.com>
发送时间：2021年12月10日(星期五) 04:27
收件人：zhenyu.ren <zhenyu....@aliyun.com>
抄 送：lttng-dev <lttng-dev@lists.lttng.org>
主 题：Re: 回复：[lttng-dev] lttng-consumerd can NOT get notification to consume ring
buffer data
----- On Dec 7, 2021, at 8:47 PM, zhenyu.ren <zhenyu....@aliyun.com> wrote:
Hi, Desnoyers
For various reasons, none of the three suggestions you mentioned about
reproducing the problem can be achieved easily.
I am also wonder why commit counter not updated...It seems that the span
producers not complete commiting(be killed?hang?)? It‘s very hard to debug all
span producers since they are too many...
So, Can I make lttng-consumer just skip out this subbuffer and try to
consume the next subbuffer?
No, it's not that simple. Once this state is reached (e.g. caused by a killed
application), just skipping this sub-buffer
will not allow applications to continue using this buffer for tracing until it
is destroyed.
Thanks,
Mathieu
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
_______________________________________________
lttng-dev mailing list
lttng-dev@lists.lttng.org
https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
