Would have been helpful to know if the initial post was referring to Version 6 also. In Version 7 data gathering for the network graphs works different, so the peaks should be an relict of the past.
Would have been helpful to know if the initial post was referring to Version 6 also. In Version 7 data gathering for the network graphs works different, so the peaks should be an relict of the past.
Andreas, I'm running 6.303, and this is quite annoying.
Our ISP runs mrtg on our internet connection, but theirs was down a few days ago, so I tried to use ASL's, and ALL of the graphs (ethernet, IPSEC, LO; current and past) were useless due to the spikes.
Upgrading our production env. to v7 will not be an option for at least several months, for numerous reasons.
Thanks Ian - I should check posting dates first. I am surprised there really are people using the search function before posting a new thread [;)]
I agree that this is an annoying bug. We've invested quite some time on this bug already and made a couple of attempts to solve it, which always only reduced the chance of peaks showing up. It is definitely not something that can be squashed easily.
In the last months our main focus was making V7 as fast and smooth as possible, and it is more likely that we will concentrate on fixing those "numerous reasons" you mentioned than addressing an issue with small operational impact on an old version. I won't say it will never happen - I don't create any schedules or timetables, and I do not set priorities. This is just my personal view which I am sharing with you as fellow techies.
