Weird BGP path selection problem
Hi, all,
I am seeing a weird BGP path selection problem on 4948 switch running cat4500-entservicesk9-mz.122-46.SG.bin code, this switch has two uplinks to the same ISP's different edge router, one circuit is primary the other one is strict backup, only default route is accepted from ISP. I am setting both local preference and weight to the default route advertised over backup link, however neither one is taking effect, BGP still thinks the backup link is better, what could be wrong?
rtr#sh ip bgp 0.0.0.0/0
BGP routing table entry for 0.0.0.0/0, version 105
Paths: (3 available, best #2, table Default-IP-Routing-Table, not advertised to EBGP peer)
Not advertised to any peer
17675, (received & used)
203.169.8.37 from 203.169.8.37 (61.211.160.150)
Origin IGP, localpref 100, valid, external
Community: 65001:0 no-export
17675
203.169.8.45 from 203.169.8.45 (61.211.160.151)
Origin IGP, localpref 90, weight 90, valid, external, best <====
Community: 65001:0 no-export
17675, (received-only)
203.169.8.45 from 203.169.8.45 (61.211.160.151)
Origin IGP, localpref 100, valid, external
Community: 65001:0 no-export
Thanks
Hi,
On cisco routers , weight is having highest preference to decide best path. By default for received route, weight is 0 but you are setting weight 90 to backup path and that is why it is getting preferred (higher is better). Please remove weight and let local preference be 90 (lesser than route on primary path)
--Pls dont forget to rate helpful posts--
Regards,
Akash
Similar Messages
-
BGP Path Selection - Favor Oldest Routes
I've been poking around in a few test routers trying to find where BGP states how long a route has been known from a neighbor. Based on Cisco's BGP path selection article: http://www.cisco.com/c/en/us/support/docs/ip/border-gateway-protocol-bgp/13753-25.html, #10 states BGP prefers the oldest known route.
What command shows the amount of time a route has been known via BGP?Thanks for your reply, Paul.
The first command just shows the same timer as the sh ip bgp summary timer. It's just the timer of the neighbor relationship.
The second command just displays how long the route has been in the routing table. I've tested this and found that when BGP loses a route to a network and then selects a different path that it had known about, the timer resets to 0. Even though it had known about the path for a while, it still resets to 0.
So thanks to everyone for your responses, but I'm still looking for some way to see the age of a BGP-learned route. -
With reference to cisco's document on BGP Best Path Selection Algorithm (http://www.cisco.com/c/en/us/support/docs/ip/border-gateway-protocol-bgp/13753-25.html).
Out of given 9 paths why 6th has been selected even though AS_PATH for 8th route is better.
Can anyone explains here, as this document has not considered the AS-PATH during path selection and used lowest ROUTER ID only.
Thanks in advance and expect technical explanation here.Hey Buddy
The AS_PATH for both is only 1, don't get confused by (AS_SET) which only counts as 1 no matter how many AS are in the set. Refer to section "How the Best Path Algorithm Works"
4.Prefer the path with the shortest AS_PATH.
Note: Be aware of these items:
◦An AS_SET counts as 1, no matter how many ASs are in the set.
So bearing the above in mind
Example: BGP Best Path Selection
Path6
(64955 65003) 65089 --- this equals 1
172.16.254.226 (metric 20645) from 10.57.255.11 (10.57.255.11)
Origin IGP, metric 0, localpref 100, valid, confed-external, best
Extended Community: RT:1100:1001
mpls labels in/out nolabel/362
!--- BGP selects this as the Best Path on comparing
!--- with all the other routes and selected based on lower router ID.
Path8
(65003) 65089 --- this equals 1
172.16.254.226 (metric 20645) from 172.16.254.234 (172.16.254.234)
Origin IGP, metric 0, localpref 100, valid, confed-external
Extended Community: RT:1100:1001
mpls labels in/out nolabel/362
Comparing path 6 with path 8:
Both paths have reachable next hops
Both paths have a WEIGHT of 0
Both paths have a LOCAL_PREF of 100
Both paths are learned
Both paths have AS_PATH length 1 --- because the (AS_SET) always equals 1
Both paths are of origin IGP
Both paths have the same neighbor AS, 65089, so comparing MED.
Both paths have a MED of 0
Both paths are confed-external
Both paths have an IGP metric to the NEXT_HOP of 20645
Path 6 is better than path 8 because it has a lower Router-ID.
Hope it helps (: -
cs4
weird buggy issue; when i create overlapping paths with the pen tool, then make a selection and a layer mask from the selection, the overlapped parts only get masked SOMETIMES. i just did one where the path had several paths inside other paths (but never a path inside a path inside a path ) and some of the "holes" got masked and are fine but some of them did not, i had to direct select the paths and make another selection and then paint on the mask. it's no big deal, i can work around it... but i dont want to. it takes longer. any ideas? i am pretty sure that i had the "exclude overlapping areas" option selected the whole time.... pretty sure....mostly.... like 90% sure... i hope i did... 30%?I think that you need to make SURE you have all the path points selected before selecting the appropriate mode for intersecting paths.
It sounds like you know how to do this, but for newer users:
• With the Pen Tool, CMND Drag around all the paths so that all points are selected.
• In the Menu Bar for the Pen Tool, on the right, you'll see 4 icons indicating overlap/intersect options. Usually the mode on the right is the one you want, (Exclude Overlapping Path Area) but in some cases the other modes are desired.
• Now load the path as a selection.
On the other hand, I find all kinds of buggy things occur with paths, especially if they are copy/pasted from another file.
Also, Vector masks behave very weirdly. In CS4 it is only possible to load the Vector Mask as a selection if that is your active layer. In order to load a different path as a selection, you have to make a layer WITHOUT a Vector Mask active, then CMND click the Path you want. Go figure.
Craig -
hi,
i have the following cli show command output,
R2#show bgp ipv4 unicast
BGP table version is 11, local router ID is 192.168.220.252
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete
Network Next Hop Metric LocPrf Weight Path
* i192.168.30.0 192.168.110.70 0 100 0 63000 i
*> 192.168.220.70 0 0 63000 63000 i
* i192.168.40.0 192.168.110.70 0 100 0 63000 63000 i
*> 192.168.220.70 0 0 63000 i
R2#
why isn't the route through the shortest AS path not selected as the best route for 192.168.30.0. ?
thanks,
uddikaR2#
R2#
R2#show ip bgp 192.168.30.0
BGP routing table entry for 192.168.30.0/24, version 7
Paths: (2 available, best #2, table Default-IP-Routing-Table)
Advertised to update-groups:
2
63000
192.168.110.70 (inaccessible) from 192.168.111.251 (192.168.111.251)
Origin IGP, metric 0, localpref 100, valid, internal
63000 63000, (received & used)
192.168.220.70 from 192.168.220.70 (192.168.220.70)
Origin IGP, metric 0, localpref 100, valid, external, best
R2#
R2#
thanks, i noticed that R2 does not have the route for the next hop, 192.168.110.70. -
BGP Best Path Selection Algorithm
How is the administrative distance positioned in the bgp route decision ?
i.e If a route is learned from iBGP with higher "local prefernece" and eBGP with lower "local prefernece" - which path will be installed in the routing table
the path learned from eBGP or the path with higer local prefernce ?For your scenario the path with the higher local pref will be installed in the routing table althogh its ibgp.
if a router recieves the same prefix from 2 neighbors 1 from ibgp and the other from ebgp
the router will compare them with the bgp path selection algorithm
the one that wins will be installed in the routing table with the admin distance of the kind of route it is so if the ibgp route won the path selection you will see in the routing table the admin distance of 200,if the ebgp route won you'll see 20 in the admin distance.
so remember the ibgp/ebgp comparision is the 9th in the path selection algorithm so an ibgp route can win the path selection by (local pref weight....)
and if the ibgp won then you'll see the ibgp admin dstance in your routing tables. -
Specific path selection in E-BGP
I have two routers ASR 9K platform with the image file is "disk0:asr9k-os-mbi-4.3.4.sp4-1.0.0/0x100305/mbiasr9k-rsp3.vm"
Primary link b/w Router 1 and 2 :
Router 1 ------> Directly connected with 30G to Router 2 on a bundle ( Neither ISIS nor BGP running on this link )
Router 2 ------> Directly connected with 30G to Router 1 on a bundle ( Neither ISIS nor BGP running on this link )
Secondary link b/w Router 1 and 2 :
Router 1 and 2 is connected on TenGig /0/0/0/4.451 and I am using secondary link for both Internet and Private (VPN) traffic as this link is running ISIS as my IGP and IPv4/v6 Unicast and IPv4/v6 Labeeled Unicast Peering using this interface
My query is how can I seggregate my IPv4/v6 Unicast Traffic in secondary link and IPv4/v6 Labelled Unicast traffic in my Primary link
Please suggest how can I do in BGP to select one path for Internet Prefixes and another for Private Prefixes
Many thanks in advance
Sankar.Hi,
The OSPF traffic would not pass through the VSL link. The path would directly go from each 4500 to the 3945 (Equal cost load balancing). I think, the 3900 series supports Etherchannel, if this is the case you can also create a L-3 Portchannel between the VSS and 3945 router. This way you use one /30 instead of 2 and you still have redundancy. For BGP, I would do one peering with Loopbacks.
HTH -
Inject BGP Default Routes into Multiple VRF before Best Path Selection
Hello,
I have the following setup:
Multiple Border Routers with eBGP sessions to external AS. We receive a default route from this multiple AS to keep the Table manageable. We noticed an important part of our traffic was been SW routed instead of CEF when we had the Full Internet table. Router Resources came to the ground when we changed to a default.
Now I want to separate this default routes into different VRF. Attached is the Diagram.
My question is, the multiple default route all go into the BGP Table. The BGP table then select the best route and place it on the RIB and then to the FIB.
I want to redistribute the different Route on the BGP table prior to the Best path selection algorithm and placed on the RIB.
How can I achieve this?Hi,
Redistribution of multiple routes to same prefix is not possible. Even if you have configured BGP multipath and all different bgp routes got installed into routing table, during redistribution only route will be redistributed.
Also would like to understand the requirement of redistributing multiple BGP routes in to IGP. As per your diagram, 3 different eBGP sessions are on three different routers, so you can prefer eBGP route over iBGP received from other routers and can distribute eBGP route to IGP from each router. Thus you will have three different default routes in to IGP in core.
Please don't forget to rate this post if it has been helpful
- Akash -
DMVPN + MPLS best-path selection
Dear Community
We're in the process of deploying DMVPN as a backup solution to MPLS. All that is working great!
The DMVPN wan is dual-cloud, with 2 hub routers in each cloud. Phase 3 (nhrp shortcut) is enabled on all the spokes.
For routing, all the customer subnets are advertised in MPLS, whereas for DMVPN hub advertises only a summary to 10.0.0.0/8. The protocol for both is BGP. For DMVPN, the hub routers resides in one AS (65002) and all the spokes another common AS 65102. DMVPN is therefore peered eBGP hub > spoke.
For customers connected to MPLS, the DMVPN serves as backup only solution. Best-path selection by longest prefix match.
We have other customers coming on board who wish to join the same WAN but don't have the $$$ for MPLS so are opting for DMVPN only.
Now, I have a requirement to enable spoke-to-spoke for a DMVPN only site (spokeA) to an MPLS site (spokeB). The problem is it doesn't seem to work properly as the hub router sees the best path to spokeB site via MPLS, not via DMVPN. The spoke-to-spoke is never formed, and remains spokeA > hub > mpls > spokeB. The return path is better = spokeB > DMVPN > hub > spokeA (this is because spokeB sees no route from MPLS for spokeA, so follows 10.0.0.0/8) route.
I look for any feedback that can help to meet this requirement?
And if any advice on the general design would be really appreciated.
Thanks a lot!
PhilPhil,
I did a short lab around this ... wanted to make sure I'm not saying something stupid.
While I can't claim it's the _optimal_ solution for your setup it seems to work in my lab.
Spoke1 LAN 192.168.101.0/24 (AS 65001)
Spoke2 LAN 192.168.102.0/24 (AS 65002)
HUB LAN 192.168.111.0/24 (AS 65000)
192.168.1.0/24 DMVPN subnet.
A single (i)VRF - DMVPN exists on hub, only and is assigned only to DMVPN tunnel interface.
Excuse a few hacks a had to use... default routed via default-originate for example :-)
Hub
R10-P#sh run int tu0
Building configuration...
Current configuration : 281 bytes
interface Tunnel0
vrf forwarding DMVPN
ip address 192.168.1.1 255.255.255.0
no ip redirects
ip nhrp map multicast dynamic
ip nhrp network-id 1
ip nhrp shortcut
ip nhrp redirect
tunnel source Loopback0
tunnel mode gre multipoint
tunnel protection ipsec profile PRO
end
R10-P#sh run | s r b
router bgp 65000
bgp log-neighbor-changes
network 192.168.111.0
redistribute static
neighbor 10.112.112.1 remote-as 65001
neighbor 10.112.112.1 route-map SPOKES_MPLS in
default-information originate
address-family ipv4 vrf DMVPN
neighbor 192.168.1.101 remote-as 65001
neighbor 192.168.1.101 activate
neighbor 192.168.1.102 remote-as 65002
neighbor 192.168.1.102 activate
exit-address-family
R10-P#sh run | s vrf defini
vrf definition DMVPN
rd 1:1
route-target export 100:1
route-target import 100:1
address-family ipv4
import ipv4 unicast map DEFAULT
export ipv4 unicast map SPOKE_SUBNETS
route-target export 100:1
route-target import 100:1
exit-address-family
address-family ipv6
route-target export 100:1
route-target import 100:1
exit-address-family
Result on spoke
R1-PE#traceroute 192.168.102.1 source e2/0
Type escape sequence to abort.
Tracing the route to 192.168.102.1
VRF info: (vrf in name/id, vrf out name/id)
1 192.168.1.1 [AS 65000] 5 msec 10 msec 2 msec
2 192.168.1.102 [AS 65000] 4 msec * 5 msec
R1-PE#traceroute 192.168.102.1 source e2/0
Type escape sequence to abort.
Tracing the route to 192.168.102.1
VRF info: (vrf in name/id, vrf out name/id)
1 192.168.1.102 [AS 65000] 6 msec * 6 msec
routing on hub
(sanitized)
R10-P# sho ip route
Gateway of last resort is 10.100.100.2 to network 0.0.0.0
S* 0.0.0.0/0 [1/0] via 10.100.100.2
10.0.0.0/8 is variably subnetted, 13 subnets, 2 masks
B 192.168.101.0/24 [20/0] via 10.112.112.1, 00:06:40
B 192.168.102.0/24 [20/0] via 192.168.1.102 (DMVPN), 00:00:03
192.168.111.0/24 is variably subnetted, 2 subnets, 2 masks
R10-P# sho ip route vrf DMVPN
Routing Table: DMVPN
Gateway of last resort is 10.100.100.2 to network 0.0.0.0
B* 0.0.0.0/0 [20/0] via 10.100.100.2, 00:06:40
192.168.1.0/24 is variably subnetted, 2 subnets, 2 masks
C 192.168.1.0/24 is directly connected, Tunnel0
L 192.168.1.1/32 is directly connected, Tunnel0
B 192.168.101.0/24 [20/0] via 192.168.1.101, 00:06:40
B 192.168.102.0/24 [20/0] via 192.168.1.102, 00:06:25 -
Path Selection – Leaving blank holes
Hello,
I'm missing a very important feature since I switched from CS4 to CS 5.5: In Photoshop CS4, when you created a path in another and selected the path, the inner path was left blank (Creating an "o" and selecting it got you the selection in shape of an o).
Now, as I'm using CS5.5, Photoshop totally ignores those inner pathes and makes one big selection (example: http://dl.dropbox.com/u/5024014/path.png). This is very bad for me, because I have to put big industrial machines on transparent background, and now I have to create dozends of paths for every litte hole and then must put them together manually, which takes a lot of time.
So, if anyone knows how to solve this problem and getting my path selection back to how it worked in CS4, I would be very glad!
Greetings, Andreas -
Path Selection between 10 gig fiber and microwave
Hello everyone,
my network is running OSPF as an IGP, i have a 10 gig Ethernet fiber connected between two sites and a microwave link as a redundant connection.
since ospf metric is cost ( or bandwidth ), the 10 gig ethernet connection is always preferred. however, sometimes the 10 gig link is flapping or the bit error rate is bad, is there anyway to change the path selection to go through the microwave when the bit error rate in the 10 gig link is bad or the link flaps ?
basically can we make the path selection based on anything than the speed or cost ?Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
Bandwidth can be a metric to OER/PfR. Much else can be used by OER/PfR.
The intent of this technology is sort of described by the names, Optimized Edge Routing (v1) and Performance Routing (v2).
Both can account for path bandwidth and/or analyze performance.
Understand typical dynamic routing protocols keep track of paths between source and destination and some have a way to "weight" paths (for example, OSPF link cost [which by RFC, hasn't nothing to do with bandwidth, but is often based on that]).
OER/PfR, for example, can run their own SLA tests.
Years ago, I set up OER in large dual MPLS/VPN environment. Our initial "problem", after activation, our WAN performance monitoring tools (and our users!) no longer "saw" any WAN performance issues. They were still happening, but OER "saw" them first, and worked around them before the monitoring tools saw them. -
After reading up on some of the problems people are having with the path/shape changes in CC 2014, I haven't seen an exact mention of the problem I am having so here I am.
2 part problem really:
1. When using the Path Selection Tool or Direct Selection Tool, sometimes I am unable to switch to any other tool using a keyboard shortcut, regardless of isolation mode or Active/All layers toggles. I press Z or B or whatever, and nothing happens. I have to click the tool icon to get it to work. This seems to happen randomly.
2. When using the Path Selection or Direct Selection Tool, after a few operations of moving points around, when I move to another layer to work on another shape the shape outline and the handles are not visible. When in All Layers mode, the layer is switched to the new shape as expected and the Properties panel shows details of the shape path but I can't see the outline. The only way I have found to 'fix' this is to draw a new shape or reopen the document or Photoshop itself.
Are these 'problems' just me missing some new way of working brought into the shape tools or is it as borked as it seems?
ThanksNo I have been using CC for about 8 months or so. I updated to CC 2014 a couple of days ago. I have been using the shapes with no issues up till now. The changes to the shape system seemed to bring it closer to illustrator, so it was logical. A change from the past, but I thought I was up to speed with it.
As for keystrokes to reproduce the problems I am having - that is part of the problem it seems kind of random as to when it happens.
When working on a shape, pressing return will turn the outline on and off as expected. Pressing esc will turn it off as expected, double clicking will take it into isolation mode as expected but then sometimes, the whole outline becomes unresponsive and it isn't visible. The shape layer is selected and the layer selection changes as expected when I have All Layers mode on and I click different shapes. The path is also selected in the Path panel. It shows up in the properties window too. Clicking/double clicking the shape, return and esc make no difference. I don't get it its all happening randomly though
Then I have the 'unable to switch tools' problem where shortcuts for different tools stop working. I have to click the tool icon to switch. Drawing a new shape brings it all back As I said in my first post, it could just be me missing something but i'm stumped. At this point i'm going to try a reinstall but I don't think it'll make much difference with something like this.
Thanks for the reply though. -
Hello Can Someone Help Me Im Having Problems with my paths panel it seems the path selection thumbnail has a circle and slash through it.. when i try and load it as a selection i cant.
Heres the screen shot the circle with the slash isnt highlighted because im hitting the prt sc button but all of the thumbnails in the path panel has that circle and slash
-
Why can't I make a selection with my path selection tool?
While following a tutorial, I am now unable to make a path selection. I know how to select a path, have done it before, but now my path selct tool does not select a path. Did I change settings or something else that I am not aware of? Any help would be appreciated.
A changed setting may not be your problem.
One pertinent question is: Can you SEE the path you are trying to select ? That is, do you see the thin line that Photoshop uses to show you the locus of points that define your path. This is distinct from the stroke that you may have applied to the path and distinct from the fill which you may have put within your path. The point being that if you cannot see the path you cannot select it. If you cannot see the path go to the paths panel and left click on the path of interest. This will show you the path line and then with your path selection tool active clicking on any part ofthe path line will select the path.
By the way, depending on path fills and path strokes that may be present it may be difficult to see the path line.
Paulo -
I cant make path selection work on the inner areas of my path selection
The problem is that it's only loading the outline of my path as the selection. All the inner areas aren't included, and I want to chop them out as well
Activate the Path Selection Tool.
Option-click on the compound Path's thumbnail in the Paths panel. That will select all components of the Path.
In the options bar, click the Subtract button to ensure all components are negative.
Click on the canvas outside of the entire path to deselect all of the path. Click on the line of the main outline path to select only it.
In the options bar, click the Add button to make the main outline positive.
The compound path should be correct now.
You can target the Path and the image layer then Add Vector Mask to the image layer, or Command-click the Path's thumbnail to get a marquee then Add Layer Mask to the image layer.
Maybe you are looking for
-
Which Adaptors/Cords do I need to hook my mid-2010 Macbook Pro into my new HDTV?
I heard they changed the port (to include audio as well as video) after my model and I want to make sure I order the right adaptors so I can hook my macbook into my new HDTVs HDMI port. Can someone point me in the right direction? Thanks, Jake
-
I got the new iPhone5 back in Dec, I hooked it up to my computer for the 1st time this wk. It deleted all new data added since Dec (notes,contacts,texts,pics, info added to apps) and reverted back to all my old data (literally uploaded all of my old
-
Query Designer : change in charactristic
Hi Experts, I have changed long text and short text of one charactristic but it is not reflecting the new name in reports.when i checked it in query designer on the left hand side under Dimensions it is showing me the new name but in Free charactrist
-
How to test the huge payload in IR in XI/PI
Hi experts, I am stuck up with the testing of the sender payload. Because of the data which is containing in sender payload is huge. Actually i am getting the error in SXMB_MONI mapping error. <SAP:Category>Application</SAP:Category> <SAP:Code area
-
I'm just starting w garageband. I have a Mac Powerbook G4, a fender electric guitar and a fender cybertwin amp. The amp has various jacks: midi in, midi out, midi thru, xlr r & l lines out, spdif output for connection to digital equipment such as a d