Re: subrate SFP?
Unless I missed something, iLO is always on when the machine has power; as such, WOL shouldn't be coming into play on reasonably modern HP servers. (That said, the power draw is likely still the reason, although I can't readily confirm Charles' observation on my only rackmount Gen8.) Jima On 2013-08-31 13:38, Joel Jaeggli wrote: WOL uses 100Mb/s, the phy draws less that way. Sent from my iPhone On Aug 31, 2013, at 10:13, Charles N Wyble wrote: On hp proliant gen8 servers with management and ilo on same port, with the server off the ports show up as 100mbps. Jimmy Hess wrote: On Fri, Aug 30, 2013 at 6:42 AM, Jamie Bowden wrote: From: Saku Ytti [mailto:s...@ytti.fi] Considering that Dell and HP at least are shipping brand new hardware with IPMI/BMC/iLO/whatever management ports that can only speak 100mbit when every other Ethernet interface in the box at least gigabit, having a useful way to talk to that port without having to keep separate switching hardware around would be nice. I'm not holding my breath, but you know, along with a pony, this would be nice. Eh? That may have been the case a few years ago, but HP ILO4 and iDRAC7 specifically list 10/100/1000 even when using in dedicated port mode. And even in prior versions, you could have the port linking up at 1Gbps, by operating the management in Shared port mode (Sharing the management with the server's Eth0). I expect over time: support for linking up at 10/100 will get rarer and much more expensive. The niche status a 10/100 media converter as an SFP would have if produced is likely to mean it would retail at $2000+ per port device. It probably just makes more sense to go find an old obsolete top of rack switch, like a Cat3750 to get the small fraction of legacy copper ports required for out of band network and server management, which: by the way, should be part of a separate switching infrastructure anyways, to increase the chance it stays operational and useful for troubleshooting, in the event the production network experiences outage or has other issues requiring diagnosis. Jamie -- -JH -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
Re: couldn't get address for 'w.au': no more ,
On 8/31/2013 4:05 PM, Mr. James W. Laferriere wrote: Hello All , Are the roots for .au lost in the haze someplace ? FWIW, I'm seeing records for both w.au and z.au from here: w.au. 172800 IN A 37.209.192.5 ;; Received 38 bytes from 2001:dcd:4::5#53(2001:dcd:4::5) in 108 ms z.au. 172800 IN A 37.209.198.5 ;; Received 38 bytes from 2001:dcd:4::5#53(2001:dcd:4::5) in 107 ms
Re: subrate SFP?
Ah, I needed *another* reason to murder WOL in it's sleep. Thanks! Nick On Sat, Aug 31, 2013 at 3:38 PM, Joel Jaeggli wrote: > WOL uses 100Mb/s, the phy draws less that way. > > Sent from my iPhone > > On Aug 31, 2013, at 10:13, Charles N Wyble > wrote: > > > On hp proliant gen8 servers with management and ilo on same port, with > the server off the ports show up as 100mbps. > > > > Jimmy Hess wrote: > >> On Fri, Aug 30, 2013 at 6:42 AM, Jamie Bowden wrote: > >> > From: Saku Ytti [mailto:s...@ytti.fi] > >>> Considering that Dell and HP at least are shipping brand new hardware > >> with > >>> IPMI/BMC/iLO/whatever management ports that can only speak 100mbit > >> when > >>> every other Ethernet interface in the box at least gigabit, having a > >> useful > >>> way to talk to that port without having to keep separate switching > >> hardware > >>> around would be nice. I'm not holding my breath, but you know, along > >> with > >>> a pony, this would be nice. > >> > >> Eh? That may have been the case a few years ago, but HP ILO4 and > >> iDRAC7 specifically list 10/100/1000 even when using in dedicated > >> port > >> mode. > >> > >> And even in prior versions, you could have the port linking up at > >> 1Gbps, > >> by operating the management in Shared port mode (Sharing the > >> management > >> with the server's Eth0). > >> > >> I expect over time: support for linking up at 10/100 will get rarer > >> and > >> much more expensive. > >> > >> > >> The niche status a 10/100 media converter as an SFP would have if > >> produced > >> is likely to mean it would retail at $2000+ per port device. > >> > >> > >> It probably just makes more sense to go find an old obsolete top of > >> rack > >> switch, like a Cat3750 to get the small fraction of legacy copper > >> ports > >> required for out of band network and server management, which: by the > >> way, should be part of a separate switching infrastructure anyways, > >> to > >> increase the chance it stays operational and useful for > >> troubleshooting, in > >> the event the production network experiences outage or has other issues > >> requiring diagnosis. > >> > >> > >> > >>> Jamie > >> > >> -- > >> -JH > > > > -- > > Sent from my Android device with K-9 Mail. Please excuse my brevity. > > > >
couldn't get address for 'w.au': no more ,
Hello All , Are the roots for .au lost in the haze someplace ? During my attempts to reach http://www.coker.com.au/bonnie++/ I tried a 'dig www.coker.com.au +trace' Did some roots change recently ? Tia, JimL Which yielded ... ; <<>> DiG 9.9.3-P2 <<>> www.coker.com.au +trace ;; global options: +cmd . 7424IN NS i.root-servers.net. . 7424IN NS m.root-servers.net. . 7424IN NS k.root-servers.net. . 7424IN NS a.root-servers.net. . 7424IN NS h.root-servers.net. . 7424IN NS d.root-servers.net. . 7424IN NS c.root-servers.net. . 7424IN NS l.root-servers.net. . 7424IN NS e.root-servers.net. . 7424IN NS g.root-servers.net. . 7424IN NS j.root-servers.net. . 7424IN NS b.root-servers.net. . 7424IN NS f.root-servers.net. . 517152 IN RRSIG NS 8 0 518400 2013090700 2013083023 49656 . lWj707jP5hxvgq8BwU5+IVeyuE/p3wcEmuQRfzuneoFClny1L/xyaT53 IkhG57jFzRPsXbuvOM6J/9tZzkbyuN20b5T0QLuxJVQsZT20pzWSIZ54 MVcVd2HTRtq+Gr0OetDI3THRkgK06IVH0yyKrPqDCQI/iHbc+iljg21f lmc= ;; Received 857 bytes from 50.0.96.199#53(50.0.96.199) in 195 ms au. 172800 IN NS z.au. au. 172800 IN NS y.au. au. 172800 IN NS x.au. au. 172800 IN NS w.au. au. 172800 IN NS v.au. au. 172800 IN NS u.au. au. 172800 IN NS s.au. au. 172800 IN NS r.au. au. 172800 IN NS b.au. au. 172800 IN NS a.au. au. 86400 IN NSECaw. NS RRSIG NSEC au. 86400 IN RRSIG NSEC 8 1 86400 2013090700 2013083023 49656 . LZo++i1OBOYRDncdZe8aAuO1TaWgCWVXVc/aquFb0oT0LBNAbkPljT55 dQV8jlrsZyZ0QbAm09P29wuq1UBuca6a1YX72DZrvfDeqX+1oXaAlEPd ZfFl2eQsao39AZPlRVfVVw18am5VX8V4K/VmYgBeq1lmV52OVqYz2UVB ygQ= dig: couldn't get address for 'z.au': no more -- +--+ | James W. Laferriere | SystemTechniques | Give me VMS | | Network&System Engineer | 3237 Holden Road | Give me Linux | | bab...@baby-dragons.com | Fairbanks, AK. 99709 | only on AXP | +--+
Re: subrate SFP?
WOL uses 100Mb/s, the phy draws less that way. Sent from my iPhone On Aug 31, 2013, at 10:13, Charles N Wyble wrote: > On hp proliant gen8 servers with management and ilo on same port, with the > server off the ports show up as 100mbps. > > Jimmy Hess wrote: >> On Fri, Aug 30, 2013 at 6:42 AM, Jamie Bowden wrote: >> From: Saku Ytti [mailto:s...@ytti.fi] >>> Considering that Dell and HP at least are shipping brand new hardware >> with >>> IPMI/BMC/iLO/whatever management ports that can only speak 100mbit >> when >>> every other Ethernet interface in the box at least gigabit, having a >> useful >>> way to talk to that port without having to keep separate switching >> hardware >>> around would be nice. I'm not holding my breath, but you know, along >> with >>> a pony, this would be nice. >> >> Eh? That may have been the case a few years ago, but HP ILO4 and >> iDRAC7 specifically list 10/100/1000 even when using in dedicated >> port >> mode. >> >> And even in prior versions, you could have the port linking up at >> 1Gbps, >> by operating the management in Shared port mode (Sharing the >> management >> with the server's Eth0). >> >> I expect over time: support for linking up at 10/100 will get rarer >> and >> much more expensive. >> >> >> The niche status a 10/100 media converter as an SFP would have if >> produced >> is likely to mean it would retail at $2000+ per port device. >> >> >> It probably just makes more sense to go find an old obsolete top of >> rack >> switch, like a Cat3750 to get the small fraction of legacy copper >> ports >> required for out of band network and server management, which: by the >> way, should be part of a separate switching infrastructure anyways, >> to >> increase the chance it stays operational and useful for >> troubleshooting, in >> the event the production network experiences outage or has other issues >> requiring diagnosis. >> >> >> >>> Jamie >> >> -- >> -JH > > -- > Sent from my Android device with K-9 Mail. Please excuse my brevity. >
Re: subrate SFP?
On hp proliant gen8 servers with management and ilo on same port, with the server off the ports show up as 100mbps. Jimmy Hess wrote: >On Fri, Aug 30, 2013 at 6:42 AM, Jamie Bowden wrote: > >> > From: Saku Ytti [mailto:s...@ytti.fi] >> Considering that Dell and HP at least are shipping brand new hardware >with >> IPMI/BMC/iLO/whatever management ports that can only speak 100mbit >when >> every other Ethernet interface in the box at least gigabit, having a >useful >> way to talk to that port without having to keep separate switching >hardware >> around would be nice. I'm not holding my breath, but you know, along >with >> a pony, this would be nice. >> > >Eh? That may have been the case a few years ago, but HP ILO4 and >iDRAC7 specifically list 10/100/1000 even when using in dedicated >port >mode. > >And even in prior versions, you could have the port linking up at >1Gbps, >by operating the management in Shared port mode (Sharing the >management >with the server's Eth0). > >I expect over time: support for linking up at 10/100 will get rarer >and >much more expensive. > > >The niche status a 10/100 media converter as an SFP would have if >produced > is likely to mean it would retail at $2000+ per port device. > > >It probably just makes more sense to go find an old obsolete top of >rack >switch, like a Cat3750 to get the small fraction of legacy copper >ports >required for out of band network and server management, which: by the >way, should be part of a separate switching infrastructure anyways, >to >increase the chance it stays operational and useful for >troubleshooting, in >the event the production network experiences outage or has other issues >requiring diagnosis. > > > >> Jamie > >-- >-JH -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
Re: subrate SFP?
On Fri, Aug 30, 2013 at 6:42 AM, Jamie Bowden wrote: > > From: Saku Ytti [mailto:s...@ytti.fi] > Considering that Dell and HP at least are shipping brand new hardware with > IPMI/BMC/iLO/whatever management ports that can only speak 100mbit when > every other Ethernet interface in the box at least gigabit, having a useful > way to talk to that port without having to keep separate switching hardware > around would be nice. I'm not holding my breath, but you know, along with > a pony, this would be nice. > Eh? That may have been the case a few years ago, but HP ILO4 and iDRAC7 specifically list 10/100/1000 even when using in dedicated port mode. And even in prior versions, you could have the port linking up at 1Gbps, by operating the management in Shared port mode (Sharing the management with the server's Eth0). I expect over time: support for linking up at 10/100 will get rarer and much more expensive. The niche status a 10/100 media converter as an SFP would have if produced is likely to mean it would retail at $2000+ per port device. It probably just makes more sense to go find an old obsolete top of rack switch, like a Cat3750 to get the small fraction of legacy copper ports required for out of band network and server management, which: by the way, should be part of a separate switching infrastructure anyways, to increase the chance it stays operational and useful for troubleshooting, in the event the production network experiences outage or has other issues requiring diagnosis. > Jamie -- -JH
Re: IP Fragmentation - Not reliable over the Internet?
>>> i wonder if this is correlated with the high number of probes being >>> behind nats. > > Maybe this provides a bit of insight: > From a test last week from all RIPE Atlas probes to a single "known > good" MTU 1500 host I compared probes where I had both a ping test with > ipv4.len 1020 and ipv4.len 1502. > behind NAT probes: 12% 1020 bytes ping worked while 1502 failed > non-NATted probes: 6%"" this needs publication on your adventure game of a web site, please. it will seriously 'inform' some discussion going back and forth on ietf lists. randy
Re: IP Fragmentation - Not reliable over the Internet?
could you please test with ipv6? thanks! randy
Re: IP Fragmentation - Not reliable over the Internet?
On 30/08/2013 16:36, Benno Overeinder wrote: > On 08/30/2013 01:58 PM, Randy Bush wrote: >>> In a study using the RIPE Atlas probes, we have used a heuristic to >>> figure out where the fragments where dropped. And from the Atlas >>> probes where IP fragments did not arrive, there is a high likelihood >>> the problem is with the last hop to the Atlas probe. >> >> i wonder if this is correlated with the high number of probes being >> behind nats. > > That would be a viable explanation, although we have not tried to > fingerprint the probes to figure out if this was true. > > If we will rerun the experiments in the future, we should spent more > effort into identifying the router/middlebox that is giving the IP > fragmentation problems (drops or blocking PMTUD ICMP). Maybe this provides a bit of insight: >From a test last week from all RIPE Atlas probes to a single "known good" MTU 1500 host I compared probes where I had both a ping test with ipv4.len 1020 and ipv4.len 1502. behind NAT probes: 12% 1020 bytes ping worked while 1502 failed non-NATted probes: 6%"" hth, Emile Aben RIPE NCC
