RIPE NCC Outage Report - 29 September 2012
• 4 min read
The weekend after the recent RIPE 65 Meeting in Amsterdam, we experienced a network outage that affected a number of services. Please find below a detailed report and analysis.
Mirjam Kühne
Based in Amsterdam, The Netherlands
112
Articles
197
Likes on articles
Based in Amsterdam, The Netherlands
I wrote the articles collected here during my time as community builder of the RIPE NCC and the maintainer and editor of RIPE Labs. I have since taken on a new role serving as the Chair of the RIPE Community. You can reach my new profile via the website link below.
RIPE NCC Reaching Last /8
• 3 min read
Now we have reached the last /8 of IPv4 addresses, we have looked at the allocation and assignment rate again. At the same time we also monitored the growth in IPv6 allocations.
List of Statistics Provided by the RIPE NCC
• 11 min read
This page contains a list of statistics the RIPE NCC provides on a regular basis, together with a short description and pointers to more information.
Update on AS Path Lengths Over Time
• 5 min read
One way to determine the denseness of the Internet, or its “interconnectedness”, is to look at the path length between Autonomous Systems (ASes). A while ago we looked at that and also at differences between IPv4 and IPv6 networks. Has this changed in the meantime? See our findings below.
Update on Assigning 32-bit ASNs
• 3 min read
The RIPE NCC has been assigning 32-bit Autonomous System Numbers since 2007. In this article we're providing an update about the number of 32-bit ASNs being assigned and how many are visible on the Internet.
First Measurement Results from IPv6 Launch
• 3 min read
Like last year during World IPv6 Day, the RIPE NCC has been doing various measurements during World IPv6 Launch on 6 June. We will publish the results in a series of articles on RIPE Labs.
Update on IPv6 Address Distribution
• 5 min read
At the recent RIPE Meeting we presented some IPv6 address distribution statistics which we would like to share with a wider audience. In the article below, you can find statistics on IPv6 allocations and assignments. In particular, the article details what affect the removal of the requirement to b…
What Ever Happened to...?
• 4 min read
... those LIRs that closed. The RIPE NCC now has over 8,000 members. But how many members have closed in the meantime and why?
From 80 to 8,000 - The Growth of the RIPE NCC Membership
• 4 min read
The RIPE NCC is 20 years old and it now has over 8,000 members. In this article, we drill down a bit more into the composition of our membership: what industry do RIPE NCC members come from and which countries account for the highest numbers of members and IP resources.
“dear :how to use bgp rpki server?”
Sylow, can you please clarify your question. Is it related to the article above?
“Thanks for the post! Solution 1 solved the problem of my probe.”
Thanks for letting us know, Stefan. I am glad to hear your probe works again.
“How to do this ?”
Hi Lisa, can you please clarify your question. What exactly are you trying to do? Thank you.
“hello, I have found that there is a problem with class of CombinePcapInputFormat about the version on github,maybe the constructor function is wrong.The version 1.1 is right. Now I need to change some code to fit my application.So that I need the right source code.Could you help me? would you put the version 1.1 on github ?Or would you email me the source code of version 1.1?My email adress is yxm1348@163.com. Thanks a lot. Yimi”
Dear Yimi, the author informs me that the version 1.1 is here in source: https://github.com/RIPE-NCC/hadoop-pcap/releases/tag/1.1 and as binary in the repository as described here: https://github.com/RIPE-NCC/hadoop-pcap I hope this helps.
“I'm interested in addressing the issues that have come up on my V1 probe, but it's unclear how to go about doing so.”
Hi Owen, Can you please send this as a ticket to atlas@ripe.net so we can respond in more detail. Please also mention the 12 characters printed under the MAC address of the probe. Thank you.
“Let me suggest one other hypothesis based on experience. There appear to be two types of USB storage failures, ones that can be overcome by the unplug, remove USB stick, plug in, reinsert USB stick procedure and ones in which the USB stick itself get trashed so that it can no longer be formatted, etc. The former issue is just an annoyance, but the annoyance is cumulative: if the probe disconnects once, whomever is hosting it is likely to go to the trouble to get it back up. After several times, the motivation to bring it back up and do it quickly may diminish. The latter issue is more serious, at least for those of us who do not keep an inventory of USB sticks in the appropriate size on hand, because the process is to spend a fair amount of time determining that resetting / reloading the USB stick won't work, finding time to go to the store or order a USB stick and wait for it to arrive, and then set the probe up again. Once sure, twice maybe, but, after running through three or four USB sticks, I'd guess that the odds of a probe being reconnected go down rapidly. Suggestion: provide a way for people to record when they have applied the "corrupt file system" fix, when they have replaced a USB stick, and what size and brand they have replaced it with. Wrt the latter, the data in the article probably don't show what USB stick is in a connected probe, only what kind of stick you shipped it with.”
Hi John, Thanks for constructive thoughts and your suggestion. We keep track of failure rates for various versions of RIPE Atlas probes to a certain extend. You can see some of this anlysis in this earlier article on RIPE Labs: https://labs.ripe.net/Members/philip_homburg/further-analysis-of-ripe-atlas-version-3-probe. Ultimately the best solution might be to look at completely different hardware. This is on our roadmap but will take some time to investigate.
“Hello. I need to contact admin. Thank you.”
Please contact labs@ripe.net.
“I had also this issue: with normal reboot didn't get a IP. without USB the probe gets again a IP but plugin the USB didn't solve the issue. The probe was still "not connected". the info in SoS were coming quickly. I had to reboot later several times with the USB. Now is OK. I hope for a long time .....”
Hi Jean-Michel, thanks for letting us know. We're working on a more long-term solution. Please also see this RIPE Labs article in that context: https://labs.ripe.net/Members/kistel/ripe-atlas-countering-hardware-issues-with-better-firmware
“Great (and a good answer to the people who would like the packets to stay inside human borders) but small typo: the IETF working group on DNS privacy is DPRIVE, not DPRIV.”
Oops. Fixed. Thanks, Stéphane.
“The link to the SURFnet website is broken :(”
Hi Sander, Thanks for pointing this out. I found it and updated the link. Mirjam
Showing 44 comment(s)