w*********g 发帖数: 30882 | 1 揭露抗草甘膦转基因大豆油对人类健康造成危害或“潜在危害”
作者:陈一文 更新时间:2011-7-29
陈一文:抗草甘膦转基因大豆油草甘膦残留量严重危害健康
http://blog.sina.com.cn/s/blog_4bb17e9d0102dr58.html
印度德里的Shiram实验室对除胶后的转基因大豆粗油、精炼油及其副产品检测草
甘膦残留量发现:不同的转基因大豆油发现草甘膦残留量差别较大为0.197-0.350mg/kg
,还发现:精炼油中的草甘膦残留量高于粗油。大豆油允许的草甘膦“最高允许残留量
”: 0.05mg/kg。卫生部2005年《国家标准食品中农药最大残留限量GB2763-2005》对对
棉籽油规定的“最大残留限量”规定为0.05mg/kg,与俄罗斯对大豆草甘膦残留量标准
一致。中国对大豆油没有规定草甘膦最大残留限量,应当认为是大豆油中不准有草甘膦
残留量。卫生部2011年《食品添加剂使用标准》(GB 2760-2011)列出的“食品添加剂
”清单不包括“草甘膦”,应当应当理解为,大豆油中如果检测出有毒草甘膦残留量,
应当视为非法“添加剂”!深圳市... 阅读全帖 |
|
c****3 发帖数: 10787 | 2 有些外企大概电信的MPLS VPN嫌贵,用软件VPN走公网出国
MPLS VPN根本是第二层上的,和公网是完全不同路由,不会被墙的 |
|
U*E 发帖数: 3620 | 3 发信人: wayofflying (小破熊), 信区: Military
标 题: 冷汗:揭露转基因大豆油对人类健康造成危害或“潜在危害”
发信站: BBS 未名空间站 (Fri Jul 29 23:55:48 2011, 美东)
揭露抗草甘膦转基因大豆油对人类健康造成危害或“潜在危害”
作者:陈一文 更新时间:2011-7-29
陈一文:抗草甘膦转基因大豆油草甘膦残留量严重危害健康
http://blog.sina.com.cn/s/blog_4bb17e9d0102dr58.html
印度德里的Shiram实验室对除胶后的转基因大豆粗油、精炼油及其副产品检测草
甘膦残留量发现:不同的转基因大豆油发现草甘膦残留量差别较大为0.197-0.350mg/kg
,还发现:精炼油中的草甘膦残留量高于粗油。大豆油允许的草甘膦“最高允许残留量
”: 0.05mg/kg。卫生部2005年《国家标准食品中农药最大残留限量GB2763-2005》对对
棉籽油规定的“最大残留限量”规定为0.05mg/kg,与俄罗斯对大豆草甘膦残留量标准
一致。中国对大豆油没有规定草甘膦最大残留限量,应当认为是大豆油中不... 阅读全帖 |
|
s********n 发帖数: 1962 | 4 ☆─────────────────────────────────────☆
zher (民工.一切要和经济挂钩) 于 (Mon Feb 23 20:06:55 2004) 提到:
这个不好说,取决于运营商目前采用的技术。以前的traditional IP backbone在做
traffic engineering的时候很死板几乎没有什么真正的好的应用,所以,虽然光纤
已经到位了,但是主要的traffic仍然在primary link上面。这样导致一条线路非常忙
很多的sp的backbone能达到80%以上的流量,只有这条断了以后,traffic才会到backup
link。在这种情况下,通常的做法是不断扩容,这就导致optical market looks good
但是,随着各大SP普遍采用MPLS技术,在Traffic Engineering上面可以做到更灵活,
几条链路share traffic,还有Fast ReRoute,把以前的切换时间从1-2秒提高到了
100ms左右,以前投资的链路就可以继续使用而不需要进行扩容,大家一般都升级软件
去支持MPLS就可以了 |
|
a*****p 发帖数: 189 | 5 收到Ericsson recruiter的一个职位信息,希望对大家有用。
=================
Hi Bryan,
Can you help me get the word out...
The Ericsson IP Networking team is aggressively hiring Routing (Unicast/
Multicast/IPv6), MPLS (MPLS-TP, L/2/3 VPN, VPLS, NSR, FRR) & Routing
Infrastructure for our current and next generation solutions. This is one of
the best times to make a mark on the industry, as it migrates to a seamless
, converged, all-IP world.
http://www.ericsson.com/ourportfolio/ericsson-product-catalog/ip-networking?n |
|
s*y 发帖数: 18644 | 6 Furnace有个part “Furnace Pressure Switch”不怎么work,上面写着:
"mpl-9300-v-0.35-deact-N/O-VS"
"42-24335-02"
"DC 0203 Costa Vita 1077"
在ebay上找到一个很像的:
除了中级有个 'deact' word.
“Furnace Pressure Switch MPL-9300-V-0.35-N/O-VS”
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=190749858886
这个deact啥意思,去哪里买replacement part呢? |
|
s********e 发帖数: 255 | 7 刚好你说的这两个地方我都呆过。我觉得呢,看你喜欢什么样的风格的玩法。
mpls你可以看到很美的自然风光,但是中部地区嘛,主要还是花花草草的风景为主,
Duluth你可以看看lake superior,但是1天也足够了。所以mpls+Duluth差不多可以玩
三天吧,再久看的风景可能就重复了。
ocean city主要就是海边游泳,吃海鲜,不过这个季节可能有用还是太冷。大概能玩两
天吧。
不过maryland的话好处就在于它周围的地方你都能顺便去,DC的那些历史建筑,博物馆
,使馆区等等就够你看上两天吧。然后annapolis, alexandria等小城也都很漂亮可以
一并游览。
然后MD和VA还有几个NATIONAL PARK也不错,都可以看看。 |
|
z**r 发帖数: 17771 | 8 Metro Ethernet switching deployment bootcamp
MPLS VPNs over IP Tunnels bootcamp
Building Core Networks with ISIS and BGP plus MPLS VPNs
Cisco ASR9000 Essentials lab guide
Cisco XR12000 Essentials lab guide
etc.
Next monday I am going to recycle these if nobody picks them up :) |
|
s*y 发帖数: 18644 | 9 【 以下文字转载自 Living 讨论区 】
发信人: sky (天), 信区: Living
标 题: furnace part问题
发信站: BBS 未名空间站 (Sun Feb 17 14:00:12 2013, 美东)
Furnace有个part “Furnace Pressure Switch”不怎么work,上面写着:
"mpl-9300-v-0.35-deact-N/O-VS"
"42-24335-02"
"DC 0203 Costa Vita 1077"
在ebay上找到一个很像的:
除了中级有个 'deact' word.
“Furnace Pressure Switch MPL-9300-V-0.35-N/O-VS”
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=190749858886
这个deact啥意思,去哪里买replacement part呢? |
|
|
|
|
|
z**r 发帖数: 17771 | 14 I have a topology like
CE1----dot1q-----PE1-------MPLS cloud-------------PE2-----dot1q--------CE2
PEs run VPLS over LDP. CE1 is sending dot1q traffic with COS value 2, on the
PE1, I rewrite the MPLS EXP bit to 4 when matching COS 2. So on the CE2, what
COS value am I supposed to see? |
|
u*v 发帖数: 18 | 15 俺对QoS研究很少,我想至少在MPLS TE有用,对进来的Traffic进行分类,确认打什么标
签,走哪条路经,以后就是标记交换了,所以必须在进MPLS时就打上tag。 |
|
u*v 发帖数: 18 | 16 俺对QoS研究很少,我想至少在MPLS TE有用,对进来的Traffic进行分类,确认打什么标
签,走哪条路经,以后就是标记交换了,所以必须在进MPLS时就打上tag。 |
|
b******e 发帖数: 66 | 17 Who are they? what kind of product are they delivering or promised to deliver?
what the major problem are they trying to solve? Some startups were trying to
build such PWE3 switch boxes to interconnect ATM/FR/SONET/ETHERNET/HDLC/
blabla, set up PWs and directly hand the traffic over to MPLS core, the main
purpose of such box is to ease SP's non-interruptive migration and converge
everything to IP/MPLS core. Just wondering what is the potential of such box?
I can see some con side of such box: t |
|
z**r 发帖数: 17771 | 18 well, MPLS was orginally designed for providers, but actually many big
enterprise networks deployed MPLS to their backbone as well. large
enterprise network can be very similar to a provider network, say Boeing,
which has about over 1M ports, the network is bigger than 90% providers in
US. So is IBM, AT&T IT, etc.
and sometimes, it's not the reason of network size, instead it's the
operational issues, say, 3 companies merge together, it would be easy to
have them on separate VPNs
IPsec or
total |
|
s**********y 发帖数: 3366 | 19 SP这几年主要是炒mpls, 过去的frame-relay, ATM以经炒熟了.
MPLS也是以routing为基础的.反正不论是SP还是企业网.routing是基础.从就业看,企业
网的需求更大.必尽SP就那么几家. |
|
m********d 发帖数: 188 | 20 I would recommend a book called "mpls-enabled applications". it's device/
vendor neutral without single line of configuration, but describes very well
how mpls works. |
|
he 发帖数: 2025 | 21 64.102.x.x
楼上两位是邻居咋的。
T |
|
|
|
z**r 发帖数: 17771 | 24 you are probably right :) |
|
|
w***s 发帖数: 321 | 26 FRR Path Protection与原来SONET APS效果是类似的。
只是Cisco没有实现,貌似 Juniper有。 |
|
|
f*****n 发帖数: 646 | 28 目前的主流解决方案是MSTP, 正在向以包交换为核心的基于Ethernet/MPLS/IP等技术的
方案演进, 主流标准是MPLS-TP |
|
s*****g 发帖数: 1055 | 29 Why in interAS scenario, BGP is better than LDP? as far as VPLS is concerned
, it should not care about BGP/ASBR etc, only it needs is MPLS
transportation to remote PEs, and LDP session to remote PEs to exchange L2
FECs.The issue that how to signal MPLS LSP over AS boundary is irrelevant to VPLS signaling.
I thought only Juniper is sticking to BGP VPLS, while the rest of world uses
LDP VPLS and use BGP for auto-discovery.
One obvious advantage of BGP-VPLS is that BGP has router-reflector concept... 阅读全帖 |
|
z**r 发帖数: 17771 | 30 来自主题: EmergingNetworking版 - L2TP协议 大规模应用的可能就sprint一家,这个东西是可以和mpls互补,不过如果有了mpls,也
的确没有必要上两套transport。不光l2vpn,l3vpn也可以
以再 |
|
z**r 发帖数: 17771 | 31 RFC 5883? BFD triggered FRR isn't new to MPLS FRR, but for IP FRR triggered
by BFD, I don't know, but it's described in the RFC 5714 section 5.2.3 "
Methods for link/node failure determination may be based on techniques such
as BFD", so there should be some kind of implementations in IP FRR? at least
, it's a good match to MPLS FRR to me :) |
|
c*****i 发帖数: 631 | 32 我说的full-mesh ibgp是说core里面所有router都跑ibgp,不是说pe之间的full mesh。
"TE需不需要mpls"是问如果你要在一个core里面setup te,你需要config mpls吗。
, |
|
c*****i 发帖数: 631 | 33 full mesh bgp,bgp neigbour之间都是直连的,为啥要igp,你自己去试一下就知道了。
另外te是靠rsvp setup的和mpls ldp没有关系。我可以肯定的告诉你,而且有isp这样
用。
then
TE can establish the LSP you want, but you still need MPLS functionality to
switch the labeled packets. |
|
z**r 发帖数: 17771 | 34 MPLS-TP is very cheap comparing to mpls, a lot of backbone may go that way
也可以是Optical Switch。 |
|
a***n 发帖数: 262 | 35 Management think the proposal is cool. Next generation campus network has to
be different from what we have - 10 distribution catalyst 6500 doing
routing and security.
Current MPLS is too complex for new network engineer. MPLS might need CCIE
level engineer to design, implement and maintain.
The new design is easy to manage. Just two nodes, routing and security on
two locations, what a simple design.
Cost saving.
Actually, we don't have issue to put more fibre on campus, we already have
about... 阅读全帖 |
|
d****i 发帖数: 1038 | 36 Let me use the cisco configuration as an example to elaborate my question:
if I have:
l2 vfi VFI man
vpn id 1111
neighbor 1.1.1.2 pw-class vc4
neighbor 1.1.1.3 pw-class vc5
pseudowire-class vc4
encap mpls
interworking vlan
pseudowire-class vc5
encap mpls
The pw class vc4 will ensure the VC to 1.1.1.2 type 4, while pw class vc5
will generally make VC to 1.1.1.3
type 5. Currently Cisco routers does not support the above configuration
because it does not support
interworking vlan for vpls yet, but ... 阅读全帖 |
|
b******s 发帖数: 5329 | 37 我搞网络10年多,来美国有一年了,现在这公司pay得不怎么样,有两个IE证书,sp/
voice。最近几年主要在搞sp这块,mpls已经玩得很烂了。我其实比较想去搞搞voice
and video,一直没抓到合适机会,rs我已经玩腻了,也觉得以后这东西没什么前途。
我现在在西岸(不是ca),现在有两个offer在手里,另外还有几个在谈。A是本地一个
consulting公司,金牌,找个主要做cisco voice的选手,当然什么都得做,给120k,
好处是不用搬家,公司不大,100号人吧,环境一般,具体公司生意怎么样,现在看不
出来。B在费城,也给120k,但还是做mpls而且主要用Junos,公司很大,几万人吧,他
们给relocate费。想听听大家意见。我觉得两个都开得低了点,不过最近行情不是很好
。原来谈过两个比这个好不少的,什么都谈好了,但是两个公司最后都发现今年只能要
公民或绿卡而放弃,一个也是consulting公司,另一个碰巧那职位有这要求。另我不考
虑绿卡的事情,就想赚点钱,现在用加拿大TN。我现在这地方,觉得工作机会挺少的,
东岸好象机会挺多的,我最近面试的职位,80... 阅读全帖 |
|
f*****m 发帖数: 416 | 38 WAN/MPLS的支持.
Cat6k 一开始的时候连native MPLS都没有 |
|
z**r 发帖数: 17771 | 39 IntServe
- RSVP
DiffServe
- Classification, IPP, COS, DSCP, IPv6 TC, MPLS
- Marking
-Explicit marking
-Implicit marking, policing set, trust domain, mpls modes, etc.
- Congestion management, PQ, CQ, LLQ, CBWFQ, scheduler, etc.
- Congestion avoidence, RED, WRED, FRED, TCP window
- Shaping and policing, token bucket, strict policer, conditional policer,
etc.
- MQC, H-QoS
- Implementation guide |
|
w***s 发帖数: 321 | 40 感觉比MPLS要好,不需要在Core起额外的协议,没有了MPLS最让人厌恶的Label的本地
有效性,也没有了不可聚合路由的要求。不过这东西在IPv6普世之后还有更广泛的应用
么?
advantage
LISP
) |
|
w***s 发帖数: 321 | 41 感觉比MPLS要好,不需要在Core起额外的协议,没有了MPLS最让人厌恶的Label的本地
有效性,也没有了不可聚合路由的要求。不过这东西在IPv6普世之后还有更广泛的应用
么?
advantage
LISP
) |
|
|
c*****i 发帖数: 631 | 43 换个话题,tony问你一个juniper mpls te autoroute的问题。我看下面的doc
http://www.juniper.net/techpubs/software/erx/erx41x/swconfig-ro
是不是说,如果我有2个router,之间有TE,ospf和bgp。在routerA上面我看一个bgp的
prefix,next hop是routerB,这样routerA就会直接用TE。
如果routerB后面还有个routerC也是ospf。这样routerA reach到routerC还是走普通
link,除非配了"tunnel mpls autoroute announce ospf" |
|
s*****g 发帖数: 1055 | 44 MPLS TE won't help, first, We don't have our own transport, second, even if
we do, MPLS TE signaling is not aware of underlying physical link
utilization.
Theatrically I don't think there is any difficulty of implementing what I
want, I just don't have the expertise to write the code. |
|
s*****g 发帖数: 1055 | 45 MPLS TE won't help, first, We don't have our own transport, second, even if
we do, MPLS TE signaling is not aware of underlying physical link
utilization.
Theatrically I don't think there is any difficulty of implementing what I
want, I just don't have the expertise to write the code. |
|
s*****g 发帖数: 1055 | 46 It means that the router at hop 7 does not support ( or not configured to
support ) ICMP extension for MPLS, it is most probably a pure MPLS core
router which does not run BGP (or at least does not have the route to the
source where traceroute packets are originated). |
|
r********t 发帖数: 338 | 47 通常ISP 都disable MPLS TTL propagation, 这样 traceroute就不会暴露MPLS core的
信息。 |
|
c******0 发帖数: 881 | 48 前天下午的电话二面,刚刚收到的onsite通知。
二面面了很多TCP和排错的问题:
1. TCP 连接建立过程,连接删除过程,序列号的具体计算方式,flow control的机制
。问题都具体到数字,并给出具体的场景要你算。
2. 客户总是收到RST包,表达一下排错思路。如果网络中有load balancer,你觉得跟
它有没有关系?可能有什么关系?load balancer在后端具体是怎么和服务器集群工作
的?
3. 为什么ISIS几乎可以直接支持ipv6 但是ospf不可以?ospfv3做了哪些改动?
答:TVL,加了type8 & 9的lsa。
4. 问了一些mpls和mpls vpn的基本概念。
另外,HR的邮件一定要填 Expected salary (include base, bonus and equity),有
没有给指点一下填多少合适?多谢~ |
|
x*********n 发帖数: 28013 | 49 我发现美国的recruiter贼噶好,还帮忙准备面试问题。
趁热打铁我记下来。
BGP部分
well know attributes,
origin里面有哪几个?
每一个attribute是怎么prefer的,越大越好还是越小越好,default value,能不能
transit
BGP neighborship 每一个阶段。
OSPF
area是针对interface还是router,type 7 LSA 在哪里可以看到(NSSA)
ospf neighborship state machine,每一个阶段。
Linux
怎么看ip table(netstat -rn)
怎么看source,dest IP(tcpdump),
怎么看mac-address table(arp -a)
MPLS
PE,P,CE概念,PHP概念,mpls lable多少个bits
2个neighborship我都没有回答好,就回答了一点点,还把exstart张冠李戴到bgp那里
了。linux那几个太紧张了,补救的还算快。 |
|
s*****g 发帖数: 1055 | 50 Correct me if I am wrong, I doubt this "overlay" is the same concept as GRE/
MPLS/MPLS-VPN kind of overlay, aka, no new encapsulation is involved.
question I have is, at what layer this "overlay" is on top of? |
|